Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Human Foods and Food-Consumption Patterns Food is the major route for human exposure to dioxins and dioxin-like com- pounds (referred to collectively as DLCs). Most DLCs in foods are contained in the lipid component of foods of animal origin. While DLC exposure through fruits, vegetables, and grains also occurs, it is thought to result primarily from the adhesion of soil to plant material. DLCs have relatively long half-lives in the human body, and the rate of elimination of these compounds from the body is inversely related to age and increased adiposity (Kreuzer et al., 1997), presum- ably on the basis of the age-associated increase in adipose tissue depots. Thus, the overall body burden tends to increase over time, even if exposure levels do not change. The ability to estimate dietary exposure and the within-population variability in exposure to DLCs is limited by the availability of appropriate data. The most recent National Health and Nutrition Examination Survey (NHANES) included testing of serum dioxin levels on a subsample of individuals and dietary measures on all participants, which should provide a clearer picture of exposures. (Dioxin measurements were added to this survey in 1999-2000, and the first data were released in January 2003, too late for inclusion in this report.) At this writing, the best available approach to estimating DLC exposure through diet is to combine information from two sources: the U.S. Department of Health and Human Ser- vices, Food and Drug Administration's (FDA) Total Diet Study (IDS), which has been accumulating information on the DLC content of selected commonly consumed food items since 1999, and the national food-consumption monitoring surveys that are conducted repeatedly by the U.S. Department of Agriculture (USDA). The consumption data selected for this report were based on actual 110
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 111 intakes rather than on disappearance data in order to obtain a more accurate perspective when the data was applied to exposure estimates. The committee commissioned an analysis based on these data; the analysis is discussed later in this chapter and in Appendix B. Aspects of current dietary patterns in the United States with regard to fat intake and to consumption of products that contribute the largest share of exposure to DLCs (i.e., meats, poultry, dairy products, and fish) are reviewed briefly below. DIETARY-INTAKE PATTERNS Overall, American adults derive about 34 percent of their dietary energy from fat and one-fourth to one-third of that is from saturated fat, which is largely animal fat. USDA survey data (1977-1978, 1989-1991, and 1994-1996) show that the average adult intake of fat in total grams increased slightly (from an average of 78.5 g/d to 83.1 g/d) between the 1989-1991 and 1994-1996 surveys (although the percent of dietary energy derived from fat decreased from 34.9 percent to 32.9 percent, a function of an overall increase in dietary energy intake). Saturated fat intake was more stable over the same time period at an average of 27.6 to 28.3 g/d (although again, the percent of dietary energy from saturated fat decreased). Fat intakes were substantially higher for men than for women (Ken- nedy et al., 1999~. The stability of the average absolute intakes, however, masks real changes in the food sources of fat over the same time period. A steady substitution of low-fat and skim milk for whole milk and yogurt occurred between 1970 and 1990, but consumption of cheese and of cream products (including sour cream, half and half, cream-based dips, and light cream) increased (Enns et al., 1997~. A trend toward more poultry and fish consumption and less red-meat con- sumption occurred over the same time period. Based on national food supply data (Putnam and Allshouse, 1993), the U.S. population per capita consumed an aver- age of 18 lb less red meat, 26 lb more poultry, and 3 lb more fish and shellfish in 1992 than in 1970. Adults (both men and women) consume more meat and fish than children, but children are major consumers of dairy products. On average, children con- sume almost twice as much milk and dairy products as do adults. Fish consump- tion is of particular interest for several reasons. One is that fish consumption has been increasing in the population, and this trend is generally a good one from a health standpoint. However, the level of DLCs in fish is variable; fish from contaminated waterways and those that receive feed that contains DLCs have higher DLC levels. Furthermore, there are relatively small subgroups of the popu- lation whose reliance on fish may be more important than that for the general population. These subgroups have the potential to have substantially higher DLC exposures than the general population.
2 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY American Indian groups in traditional fishing cultures (primarily in the North- east and northern Midwest) and Alaska Native and northern Canadian groups (Northern Dwelling populations) that traditionally consume large amounts of fish and marine mammals, and rural subsistence and sport fishers in various parts of the country, are potentially at risk. A number of rivers and lakes that are impor- tant for fishing, including the Great Lakes, are near or downstream from indus- trial sources of contamination. While data on dietary exposures to DLCs in these groups are limited, a few studies are relevant. Outside of American Indian groups, there is a paucity of data on the intakes of subsistence or sport fishers. Frequent consumption of game fish is of concern because many of these fish are particularly high sources of DLCs (Anderson et al., 1998~. Subsistence fishing is a cultural tradition in many African-American families and in Southeast-Asian immigrants. DLC Analyses of Foods Consumed Due primarily to the high cost per sample to analyze DLCs, there is relatively little information on the DLC content of the U.S. food supply. Recently however, FDA conducted DLC analyses for selected foods collected as market basket samples in the 1999-2000 and 2001 TDS. The TDS survey samples foods pur- chased by FDA personnel from supermarkets or grocery stores four times per year; one sample is collected from each of three cities from each of four geo- graphic regions. Each market basket contains similar foods purchased in the three cities in a given region. The market basket food samples are then prepared as for consumption, and the three samples of a food from a region are composited into a single sample, prepared in duplicate, and analyzed for DLCs. Representative ranges of DLC content in foods from a portion of the 2001 analysis are presented in Table 5-1 (the complete analysis is presented in Appendix B). The ranges on the table reflect the estimates of the absolute DLC levels in foods not exposure levels, which are determined by the amount of food consumed. Furthermore, the analysis was based on whole food or wet weight rather than on a lipid basis in order to maintain consistency with consumption data and to preserve data quality, although inclusion of the lipid content might have been useful in comparisons among reports. Table 5-1 shows that DLCs may be found in all foods, with a range of content within each food group. However, only in animal products, specifically red meat, fish, and dairy products, do concentrations reach more than 0.10 ppt. In general, higher DLC concentrations are associated with higher fat contents, par- ticularly with animal fats. Foods representing the upper levels of these ranges include high-fat cuts of beef, bacon, frankfurters, full-fat cheeses, fatty fish (e.g., salmon), and butter. On the other hand, within each of these categories, foods containing relatively low levels of DLCs can be selected, including lean steak (0.03 ppt), lean ham (0.04 ppt), cottage cheese (0.03 ppt), shrimp (0.06 ppt), and
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS TABLE 5-1 Dioxin and Dioxin-like Compound (DLC) Content Per Gram in Selected Food Groups from the 2001 Total Diet Study 113 Food Group DLC Content (ppt toxicity equivalents) Meat, fish, poultry, eggs Beef, pork, lamb Processed meats Fish, shellfish Poultry Eggs Dairy foods Cheese Cream/ice cream Milk Fats, oils, nuts Butter Vegetable oils Nuts Breads, cereals Breads Crackers Pasta Breakfast cereal Fruits, vegetables Fruit Vegetables 0.005-0.46 0.01-0.21 0.01-0.33 0.004-0.06 0.01-0.05 0.002-0.24 0.0001-0.06 0.0006-0.01 0.22 0.002-0.06 0.003-0.006 0.001-0.05 0.001-0.02 0.0001-0.02 0.0007-0.01 0.0007-0.01 0.0001-0.05 NOTE: Mean toxicity equivalents at nondetects = 0. SOURCE: Douglass and Murphy (2002). margarine (0.03 ppt). Fruit, vegetable, and grain products have considerably lower concentrations of DLCs, but because large volumes are consumed, these foods also contribute to total DLC exposure (Douglass and Murphy, 2002~. USDA's Agricultural Research Service conducts a nationally representative survey, formerly known as the Continuing Survey of Food Intakes by Individuals (CSFII), which has now been merged with NHANES. This survey provides im- portant information about the types of foods Americans are consuming and the quantities consumed (NCHS, 2003~. The 1994-1996 CSFII was conducted between January 1994 and January 1997. In each of the three survey years, data were collected from a nationally representative sample of noninstitutionalized individuals of all ages in the United States. The 1998 CSFII was a survey of children ages O through 9 years. It was conducted using the same sampling and dietary intake methodologies as the 1994-1996 CSFII so that the survey results could be merged to increase the total sample size for children. In both surveys, dietary intakes were collected through
4 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY in-person interviews using 24-hour recalls on two nonconsecutive days, approxi- mately one week apart. A total of 21,662 individuals provided data for the first day; of those individuals, 20,607 provided data for the second day. Each dietary recall included a record of all foods and beverages consumed in the previous 24 hours, including the gram weight of each food or beverage consumed. Data obtained from the sequential CSFIIs over the past 30 years indicate a trend in food consumption toward a lower total saturated fat intake and a lower total fat intake as a percent of energy consumed, although with a higher total energy intake. A summary of 1994-1996 CSFII intake data for foods that are most likely to contain high amounts of DLCs is presented in Figures 5-1 through 5-3. The figures demonstrate that differing age and sex groups tend to be exposed to DLCs through intakes of foods from different food groups. For example, male adolescents and men consume considerably more beef than women, with little consumed by children under 6 years of age. More fish is consumed by adults than by children, and more by men than by women. On the other hand, the average milk and dairy consumption is highest for young children; adults consume con- siderably less. The committee commissioned an analysis (see Appendix B) to estimate ex- posures to DLCs from the U.S. food supply for venous age and sex groups. The analysis utilized data from the TDS for samples collected in 2001 (which were 45 40 35 In 30 o O 25 3 20 1 5 CD 1 0 5 o 0-5 M/F 6-11 M 6-11 F 12-19M 12-19F 20&Up 20&Up M F Age Range (y) by Gender FIGURE 5-1 Beef consumption by age and sex, 1-day average intake, 1994-1996. Val- ues represent population intake averages of absolute amounts of beef consumed per indi- vidual for 1 day. Error bars are standard errors of the mean. M = male, F = female. SC)IJRCE- Fond Survevs Research Group (1997~.
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 500 450 is, 400 In o 350 250 A 200 a: 150 100 50 o 115 0-5 M/F 6-11 M 6-11 F 12-19M 12-19F 20&Up 20&Up M F Age Range (y) by Gender FIGURE 5-2 Fish consumption by age and sex, 1-day average intake, 1994-1996. Val- ues represent population intake averages of absolute amounts of fish consumed per indi- vidual for 1 day. Error bars are standard errors of the mean. M = male, F = female. SOURCE: Food Surveys Research Group (1997~. 40 - ~, 35- 30- (n o ~ 20 o 15 10 25 - o 0-5 M/F 6-11 M 6-11 F 12-19 12-19 F 20 & Up 20 & Up M F Age Range (y) by Gender FIGURE 5-3 Dairy consumption by age and sex, 1-day average intake, 1994-1996. Values represent population intake averages of absolute amounts of milk and dairy foods consumed per individual for 1 day. Error bars are standard errors of the mean. M = male, F = female. SOURCE: Food Surveys Research Group (1997~.
116 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY analyzed for 17 dioxin congeners). The TDS data were linked to intake data from the 1994-1996 and 1998 CSFII to estimate dietary exposure to DLCs, through foods actually consumed. The TDS weighting factors developed for use in esti- mating exposure based on analyte concentrations in TDS foods are based on results of the 1987-1988 USDA Nationwide Food Consumption Survey, and therefore are not likely to reflect current patterns of intake. Therefore, the analy- sis commissioned by the committee used results from the 1994-1996,1998 CSFII. Because relatively few foods were sampled for the TDS relative to the thou- sands of foods in the CSFII database, CSFII foods were grouped to estimate DLC content from the most similar analyzed food. This method generates substantial errors for individual intake estimates, but because FDA sampled the foods it considered most representative of the U.S. diet, the results should provide reason- able average DLC intake estimates. These limitations must, however, be taken into consideration when interpreting the results. The results of this merged analysis (presented in Figures 5-4 through 5-11) allow a view of the relative contnbution, within the limitations of the data, to total DLC intake from venous food groups based on actual U.S. dietary intake. The analysis did not utilize "usual intakes" (commonly used for nutrient intake levels from foods) because of the extensive variability of levels of contaminants (e.g., DLCs) in foods. Other Foods and Mixtures 10% Fats, Oils and Mixtures 1% Fruits, Vegetables and Mixtures 10% Eggs and Mixtures 4% Fish and Mixtures 4% Poultry and Mixtures'' Meat and Mixtures 35% Dairy Foods and Mixtures 30% FIGURE 5-4 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), children ages 1 through S years. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight.
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS Other Foods and Mixtures 13% Fats, Oils and Mixtures 1% Fruits, Vegetables and Mixtures 9% Eggs and Mixtures 2% Fish and Mixtures 6% Poultry and Mixtures' 7% 117 Dairy Foods and Mixtures 24% Meat and Mixtures 38% FIGURE 5-5 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), children ages 6 through 11 years. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight. Other Foods and Mixtures ~ 4o/^ Fats, Oils and Mixtures Fruits, Vegetables and Mixtures 9% Eggs and Mixtures 2% Fish and Mixtures 5% Poultry and Mixtures " 7% Dairy Foods and Mixtures 16% Meat and Mixtures 46% FIGURE 5-6 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), boys ages 12 through 19 years. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight.
118 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY Other Foods and Mixtures 12% Fats, Oils and Mixtures 2% Fruits, Vegetables and Mixtures 9% Eggs and Mixtures Fish and Mixtures 10% Poultry and Mixtures 7% Dairy Foods and Mixtures 1 1 % Meat and Mixtures 46% FIGURE 5-7 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), men ages 20 years and older. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight. Other Foods and Mixtures ~ Ro/O Fats, Oils and Mixtures 1% Fruits, Vegetables and Mixtures 8% Eggs and Mixtures 2% Fish and Mixtures 5% Poultry and Mixtures 7% Dairy Foods and 17% ixtures Meat and Mixtures 45% FIGURE 5-8 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), girls ages 12 through 19 years. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight.
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS Other Foods and Mixtures 13% Fats, Oils and Mixtures 2% Fruits, Vegetables and Mixtures 1 1% Eggs and Mixtures 3% Fish and Mixtures 12% Dairy Foods and Mixtures 1 polio Poultry and Mixtures 7% 119 Meat and Mixtures 40% FIGURE 5-9 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), women ages 20 years and older. Nondetects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight. Other Foods and Mixtures 15% Fats, Oils and Mixtures 1% Fruits, Vegetables and Mixtures 12% Eggs and Mixtures 4% Fish and Mixtures 4% Poultry and Mixtures 7% Dairy Foods and Mixtures 20% Meat and Mixtures 37% FIGURE 5-10 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), pregnant and/or lactating females ages 12 years and older. Non- detects = 0. Percentages were calculated based on mean 2-day average DLC intake/kg of body weight.
120 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY Other Foods and Mixtures 12% Fats, Oils and Mixtures Fruits, Vegetables and Mixtures 10% Eggs and Mixtures 3% Fish and Mixtures 8% Poultry and Mixtures - . 7% Dairy Foods and Mixtures 7o/^ Meat and Mixtures 42% FIGURE 5-11 Estimated percent contribution of foods to intakes of dioxins and dioxin- like compounds (DLCs), males and females ages 1 year and older. Nondetects = 0. Per- centages were calculated based on mean 2-day average DLC intake/kg of body weight. Because it was not possible to more closely define the food categories, the "fruits, vegetables and mixtures" and "other foods and mixtures" categories con- tain an array of items. Foods that fell within the general category of fruits and vegetables included raw and cooked vegetables and fruits, along with vegetable items prepared as mixtures that contained some milk, eggs, cheese, and meat. The others foods and mixtures category included food combinations and mixtures, some of which were made with animal food products. The imputation of foods in these two groups to data from the CSFII also required making generalizations about specific intakes. The data in the figures are relative exposures, with DLC nondetect levels assumed to be zero. The committee chose this reference value because it felt that, under these circumstances, this value gave the most accurate representation of the DLC intake. The common alternative methods (assuming nondetects = I/2 of the limit of detection or nondetects = the limit of detection) are more conservative, but they tend to artifactually indicate a greater contribution of substances known to be low in DLCs (grains, fruits, and vegetables) because of their relative roles in the total diet. The results of the analysis, discounting the other foods and mixtures cat- egory, suggest that DLC exposure levels to the general adult population from animal-food products and based on estimated amounts consumed by population groups, are greatest from meat, followed by dairy foods and fish. The estimates for dairy foods do not take into account the consumption of low-fat and skim milk because only whole milk was analyzed. The dairy foods estimates do, however,
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 121 include cheese and other high-fat dairy products, except for butter, which is included under fats, oils and mixtures. A discussion of DLC-intake estimation scenarios for skim versus whole-milk consumption appears later in this chapter (supporting data is provided in Appendix B). Because DLC content is strongly associated with the fat content of animal products, individuals who select lean cuts of beef, fish, and low-fat dairy products (including low fat cheeses) and who utilize lean animal products in combination foods will have considerably less exposure to DLCs than those who choose higher-fat versions of these foods. Table 5-2 shows the estimated exposure to DLCs through foods for all age and sex groups combined, that is, males and females ages 1 year and older. When age groups are combined, fish and mixtures move down in ranking because young children consume very little fish. This combined ranking, however, sup- ports the finding that most DLC intake comes from animal-based foods, particu- larly meat (other than lean cuts) and full-fat dairy foods. As previously stated, DLCs are found primarily in the fat component of these foods. Lower-fat, ani- mal-based foods may be consumed in place of high-fat selections to reduce DLC exposure and not alter nutrient intake, except for energy from fat. Dietary Intake Scenario for Meat, Poultry, and Fish The commissioned analysis also included an estimation scenario of DLC exposure through food that compared low (less than 1 oz) meat, poultry, and fish consumption with high (greater than 1 oz) consumption for each age and sex TABLE 5-2 Ranked Dioxin and Dioxin-like Compound (DLC) Sources from Foods Consumed Among All Age Groups in the U.S. Diet Food Category Percent Estimated Per Capita Contribution of Foods to DLC Intakea Meat and mixtures Dairy foods and mixtures Other foods and mixtures Fruits, vegetables and mixtures Fish and mixtures Poultry and mixtures Eggs and mixtures Fats, oils and mixtures 17 12 10 8 7 3 1 aPercentages represent 50th percentile intakes at nondetects = 0. SOURCE: Douglass and Murphy (2002).
122 Cal o sol an o sol ad Cq o Cq Cq o o an ·_4 x To ·_4 x To o Cq an _ Cq o U) ~ ~ an Em o to ~ =~ V) o V 4, .O ;^ o o' V ~ ~ ~4 ;^ ~ ca C) o o ~ A 11 ca ~ C) o lo, o A A 11 o so A 11 o ~ too ~ ~ ·0 ca ~ ~ V ~ ~ o .~ o ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ 0 ~ ~ ~ ~ ~ . . . . . . . . . . . . . . . . 00 ~ 0 0 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 00 ~ ~ ~ ~ 0 . . . . . . . . . . . . . . . . 00 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ ~ 0 oo a~ ~ ~ ~ ~ ~ oo ~ ~ ~ O O O .. .. .. .. .. .. .. ~ ~ ~ ~ ~ ~ ~ ~ ~ . . . . . . . . ~ ~ ~ ~ ~ 0 0 0 00 ~ ~ . . . . . . . ~ ~ ~ 0 0 0 0 00 00 . . . . . . . O 0 0 0 0 ~ 0 00 ~ ~ 00 ~ . . . . . . . . O 0 0 0 0 0 0 0 ~ ~ ~ ~ ~ 00 ~ ~ ~ ~ ~ ~ 00 ~ ~ ~ . . . . . . . . . . . . . . . . ~ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 0 ~ ~ 00 ~ 00 ~ ~ ~ O ~ ~ ~ ~ ~ O ~ ~ ~ ~ oo O ~ oo ~ O ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ·0 O ·0 O ·0 O ·0 O ·0 O ·0 O ·0 O ·0 O .~ . ~ ~ ·~ s~ O C R ~ ·~ R s-. + O O ca ca ~ ° v, ;>, ;~. ~ ;^ ~ ~ ·~ O O ~ ~ =4 ~ ~ O ca ca ca ca ~ ~ ~ ca ~ ~ ~ ~ ~ ~ ~ ~ ~ - ~ ~ cd cd ~ ~ ~ ~ ~ ca ca s~ o c) x o o s~ t~o. ca o ca ·= ~ ~ c) ·- o ~ .~ ~ o c) =.~ o ~ ·0 ~ u, ~ 11 c' 11 o' ~ ~ oo ~ l o ~ ~ ca ~ . .~ ~ o c) ~ o o ca ^ ~ ;^ ·O ~ . ~ ~ 8 ca 1 ¢ V, ~ ;^ G ~ ~ V) ~ ·0 C) ·= ¢ E-4 ° o · - · _4 .~ ¢ V, o ca O ~ ) ;^ CO ~ ca ca ~ ~ ~ :^ ~ ~R O ;^ ~ ~ ~ =, =, C.) O ca X .. R P~ 3 .= ° V F¢¢ E~ ca E~ o ~ ¢ z C)
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 123 group. The results are presented in Table 5-3 and supporting data is provided in Appendix B. The intake difference for each group can be summarized as a per- centage comparison of low with high toxicity equivalents (TEQ) intake. Using the assumption that nondetects = 0, for adolescent (12-19 years old) and adult (20+ years old) males, DLC-intake exposure for low-meat consumers is 34 and 45 percent, respectively, of that for high-meat consumers. For nonpregnant ado- lescent and adult females, DLC-intake exposure for low-meat consumers is 44 and 47 percent, respectively, of that for high-meat consumers. These results suggest that reducing the consumption of foods that contain animal fats can reduce DLC intake exposure. Although the analysis did not compare trimmed with untrimmed meats, poultry, and fish, it is fair to assume that trimming and discarding their excess fats will reduce DLC intake. There is an array of nutrients that can be obtained from these foods, including iron, niacin, vitamin Bit, and protein; lean cuts of meat and poultry will provide levels of these nutrients that are comparable to untrimmed versions while reducing intake exposure to DLCs. Dietary Intake Scenario for Skim versus Whole Milk Another scenario produced in the commissioned analysis was an estimation of exposure to DLCs through the consumption of skim milk compared with whole milk (3.5 percent fat). This was an area of interest to the committee because the subgroups within the population that are high-milk consumers are young children and girls who have not entered their childbearing years, and whole milk and full-fat dairy products are a source of DLCs. The intake estima- tion scenario compares the expected levels of DLC intake exposure for (1) all foods if whole milk versus skim milk is consumed, and (2) dairy foods and mixtures if whole milk versus skim milk is consumed. The results are presented in Tables 5-4 through 5-6. Using the assumption that nondetects = 0, a mean reduction in DLC intake exposure of at least 60 percent from dairy foods and mixtures can be predicted if skim rather than whole milk is consumed. However, when dairy intake is compared with intake from all foods, the mean reduction is approximately 10 to 20 percent. The results of this estimation scenario suggest that DLC intake exposure can be reduced by consum- ing skim rather that whole milk, but the reduction would be more significant if combined with a reduction in intake of other sources of DLC exposure, such as animal fats from meat, poultry, and fish.
124 o Cq a~ l Cq .5 Cq o o o ~o Cq Cq o o a~ ·_4 x ~o ·_4 o o a~ CO _. .= Cq U) ~ ~ =~ ~ ll ~ ~o ~0 R R ~C ~o ~ ~ R R >< O ·DR o 3 ~ o o o o o o ~ ~ . o o .~ ~ ~ O R . . . O ~ oc O . . . , O ~ O o ~ ~ O oc . . . . . O O O C oo ~ ~ ~ oo ~ ~ ~ ~ oo O O O O O O ......... O O O O O O O O O oo O ~ ~ ~ oo O . . . . . . . . . . . O O O O O O O O O O ~ ~ O ~ O oo ~ O ~ ~ ~ ~ O . . . . . . . . . . . O O ~ O O O O O O O ~ ~ ~ ~ oo ~ O ~ ~ ~ O ~ O ~ ~ ~ O . . . . . . . . . . . ~ ~ O O O O O O O O O oo ~ ~ O O O O O O ~ . . . . . . . . . . o 0 0 0 0 0 0 0 0 1 o O ~ . . . . . . . . . . . ~ ~ O O O O O O O O O oo ~ ~ ~ O ~ oo ~ ~ O ~ . . . . . . . . . . . O O ~ O O O O O O O ~ ~ O ~ oo O ~ O O ~ O, O o 0 0 0 0 0 0 0 0 1 o oo ~ ~ ~ ~ ~ O O 0 0~ 0 0 ...... . O O O O O O O O O oo ~ . . . O O oo . . . . . . . . . . . O O ~ ~ ~ ~ ~ ~ ~ O O O ~ ~ ~ ~ O ~ ~ oo . . . . . . . . . . . O O oo oo O ~ ~ ~ oo O O ~ ~ oo ca ca ca .~ ~ ~ ~ ~ . . X a a a . ~ ~ a ~ ~ ~ ~ ~ ~ L 0 ~ ~0 o ~ ,~ ~ _4 ~ .~ ~ 4~;) ·0 c) ~ c) o 4~;) -) ·g o o ~ 4~;) ·~ s~ ca , ·= ) ca ca ~ ~ o ~ o ·~x ~ om o ~ ·x t,.4-~ ·= S~ ca ~ O ca ~; 4~;) ) Lb .c ~ ·g om' ~' c) 1 1 ~ ~ ·x o .-o. ~ ¢ v, ~ oo x ~ ~ l .~ ~ ~ ca ~ o ~ o m :-o. ~ 0 ·~;) mo ca ;>, ¢ ~ ~ . ~ ~ 1 ~ · V) ~ mo ~ =-~ o ~ ~ ~ o := ~ o ~ ~ ¢-~ pR a ~ a ,- O V, ~ ~ ,) ~ ~ 00 ~ ·~ =0 ~ 4.;, ~ O .~ ~ ~ · ·0 ° ~ ~ C a ~ : ~ 3 ' ~ , _ ~
a is Al l Hi .a hi o o To o Hi - ~ Hi o o · - ~ K Cam K o Al Al .= U) 1 ~ U) ED ~ ~ oo ~ ~ o ~ ~ ~ ~ oo ~ - - ooo O R ~ ~ ~ O 11 ~ ~ ~ ~ ~ ~ ~ O 0~'0 3 Z ~ coo ;^ ~ ~ ~ oo =0 at == O ~ ~ =, O O O O ~ ~ ~ O ~ .~ z ~ ~ =0 .~^ O ~ ~ '-0 ~ -'0- R o ~ o - _ - ~31 ~ I ~ ~ ~ ooo ;^ o o ~ y o ~ h~ O ~ .g 8 =e ~ ~V o hn v ~o oo o ~ o o _I ~ o o o o I , o o o o I . . . o o o o ~ . . o o C 00 _I ~ O OC . . . . . o o o o o C ~ ~ ~ O I OC O I O ~ O ~ . . . . . , o o o o o C o o o o . . . . , o o o o _I ~ 00 _I ~ O . . . , o o o _I ~ ~ 00 I . . . o o o ~ oo ~ o o o o o o _I ~ ~ o o o o o o o ~ o o o o c o ~ o . . o o C _I ~ OC o ~ . . o o C ~ I I O C~ o o C C, o 1 oo _I O ~ o o C; o o o . . . . o o o _I 1 o 1 cm 1 o o Io o oo o ~ o o o o . . . . . . o o o o o o ~ ~ ~ ~ ~ ~ ~ ~ I ~ O . . . . . . . . . . . I O O 1 O O I I 00 I ~ ~ ~ 00 C, o o oo oo ~ ~ ~ ~ oo ~ o O O ~ ~ 00 ~ I ~ ~ ~ O _I _I _ ca s~ ca X O c; O ~ . ~ ~ ~ ~ R ~ ·= a a a ~ g~ ~ ~ O a .g ~0 o .= ~ .~ ~ O O 4,;, Ce R . ca O h~ O J4 R 3 O ~ ~ ca ;^ 5~ · ¢~m ~ Ra ' ,, ~ a .ca ~ ~ ~ ~ ~ ~ ~ Z [ ~ · ¢ V, 00 _I _I l _I
126 o Cal an Cal Cq ad ad o o to Cq Cq o o ad ·_4 x to ·_4 o o ad 11 _. ~ _' A be ~ . - 50 .= ~ Cq C) ski us o Em ~ 11 ca ~ C) .O z Y o R O ca .- Z . ;^ · C) X Em X O O 11 R Inca ca ~ V Z . O =~ ~ C) O O O ~ O =~ ~ C) O O o ~ ~ . O O . R ~ C ~ ;^ s~ o V o o . . . . ~ ~ O O oo oo . . O O O ~ . . O O . . O O O O . . O O O O . . O O ~ O . . O O oo ~ O O O . . . . ~ ~ O O oo ~ . . O O oo ~ . . O O . . O O O O O O O . . O O . . O O O O . . O O ~ O O O . . O O O ~ ~ O . . O O oo O . O O O O . O O ~ O O O . . . O O O O . . O O ~ ~ ~ O ~ O ~ . . . . . . . O O O O O O O O ~ O ~ O . . . . . O O O C O O O O O . . . . . O O O O O oo ~ oo O ~ O ~ O ~ . . . . . . . , O O O O O O O O ~ O ~ . . . . . . . , O O O O O O O O O O O . . . . . , O O O O O O O O . . . , O O O , O O O . O ~ ~ ~ ~ . . . . . . . . . . . O O ~ ~ ~ ~ ~ O ~ O ~ O O ~ ~ ~ ~ ~ ~ oo ~ O . . . . . . . . . . . O O oo oo oo ~ ~ ~ ~ O O O O oo oo ~ ~ ~ ~ ~ ~ O ca ~ X O ~ ~ V, ca ~ =0 ~ o _4 ca ~ e R ~ r ca X S~ ca S~ ~ ~ X ca X 4.;, ca ° ~ ~ ~0 U:, 0 S ~ . ~ ~ 0 .g o .~ C) S~ O ~0 ~ ca O ~ O ~ ·O O O ~ ~ ca ;^ ¢^ · R 8 ~ :, u R ~ R o ~ ca ct ca ,~ ~ ·0 C~ ~ ~ ~ 3 R R ~ ¢ ~ 5 ~ ~ t,.,. R O oo ~ ·~ O 0~' ~ ·~ ^ ~ · ·0 a ,~ 8 ~ 8 ~ .= ~ ~ ~ ~ ~ ~ O :~ ,7 ~ _ ¢ v, oo l ca o c)
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS CURRENT DIETARY PATTERNS Dietary Exposure to DLCs General Population 127 Trends from 1977-1978 through 1994-1995 show that there has been a decline in the consumption of whole milk, but there have been increases in the consumption of grains and meat-based mixtures (Enns et al., 1997~. Results of the most recently reported CSFII (1994-1996) show that, on average, Americans consumed 4.9 servings of meat or meat substitutes daily. Within the meat group, average intakes included 1.9 servings of red meat,1.2 of poultry, 0.8 of processed meat, 0.4 of fish, 0.4 of eggs, and 0.1 of nuts or seeds. Average intakes of other foods included 3.1 servings of vegetables, 1.5 of fruit,1.5 of dairy foods, and 6.8 of grain-based products (Smiciklas-Wright et al., 2002~. These data supports the observation from the commissioned analysis that DLC exposure through animal- based foods comes primarily from meat and meat mixtures. Sensitive and Highly Exposed Subgroups of the General Population Some life stage groups within the general population, particularly develop- ing fetuses and infants, may be more sensitive to DLC exposure because their developmental immaturity. These life stage groups also may be more highly exposed as a result of smaller body size and, in the case of infants, through breastfeeding. Certain cultural or behavioral subgroups of the population may have higher exposures to DLCs than the general population, based upon differences in food- consumption patterns. Such populations include some American Indian groups, Alaska Native and northern Canadian groups, and subsistence or sport fishers. These populations may consume greater amounts of fish known to have relatively high concentrations of DLCs due to the location where the fish are caught. The committee took these groups into consideration as it explored potential ways to reduce human dietary exposure to DLCs. In Utero Exposure. DLCs, although lipophilic, are known to cross the pla- centa during pregnancy (Holladay, 1999~. The total DLC body burden of the developing fetus has been estimated using cord blood samples obtained at birth, based on the assumption that this value represents in utero exposure. Koopman- Esseboom and colleagues (1994a) determined that concentrations of polychlori- nated biphenyls (PCB) in cord blood, collected on a cohort in the Netherlands in 1990, were approximately 20 percent of maternal plasma values, expressed on a concentration (volume) basis. In contrast to these findings, in the Michigan Ma- ternal/Infant Cohort Study, Jacobson and colleagues (1984) found that PCB lev-
28 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY els expressed on a lipid basis were equivalent between maternal plasma and cord serum samples. This contrast is because fetal plasma contains only approximately 20 percent of the lipid level of maternal plasma. Both studies, however, demon- strate that fetal exposure to DLCs is a function of the mother's body burden. Total fetal content of PCBs is comparable on a lipid basis with the mother. However, on the basis of total body weight, the relative DLC content is less because the average fraction of body weight represented by lipids in the human newborn is lower than the average lipid content of adults. Various developmental health outcomes have been described that are attrib- uted to different levels of exposure to DLCs and PCBs. Jacobson and Jacobson (2002) described impaired cognitive functioning in infants born to mothers who consumed PCB-contaminated fish from Lake Michigan. Other health outcomes to infants attributed to in utero exposure to DLCs includes poor psychomotor skills (Koopman-Esseboom et al., 1996), altered thyroid hormone levels (Koopman-Esseboom et al., 1994b), and reduced neurological optimality (Huisman et al., 1995~. The health effects of DLC exposure to infants and young children are further described in Chapter 2. A woman' s DLC body burden is an accumulation of her intake exposure to DLCs from birth, beginning with her consumption of breast milk during the first months of life and continuing up to the time that she becomes pregnant. As a result of her lifetime-accumulated exposure and the long half-life of DLCs, a woman's dietary modifications that are intended to reduce DLC intake during pregnancy will not have an impact on exposure to the developing fetus. Breastieeding Infants. The primary postpartum DLC exposure route to in- fants is through breast milk. The American Academy of Pediatrics (AAP) pro- motes breastfeeding of infants as the foundation of good feeding practices and recognizes the critical role of breastfeeding as primary in achieving optimal infant and child health, growth, and development. AAP recommends exclusive breastfeeding for the first six months of life and continuation of breastfeeding through 1 year of age (Work Group on Breastfeeding, 1997~. There are well- documented advantages of breastfeeding, not just to the infant, but to the mother, the family, and to society, including health, nutritional, immunological, develop- mental, psychological, social, economical, and environmental benefits (Work Group on Breastfeeding, 1997~. Besides being the ideal nutritional source for human infants, human milk contains growth factors that aid in the development of the neonate's immature gastrointestinal tract and in repair following insults such as common diarrhea! illness (Beret and Walker, 1999; Playford et al., 2000~; hormones and enzymes with multiple functions; and numerous immune factors that protect against many infections (Word and Adlerberth, 2000~. In addition, breastfed infants are leaner (Dewey et al., 1995) and have a somewhat lower risk of later childhood obesity (Dewey, 2003~. Psychological and cognitive develop-
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 129 ment also appears to be favorably affected by breastfeeding (Jacobson et al., 1999; Lucas et al., 1992~. A report from the U.S. Department of Health and Human Services (HHS) Office on Women's Health, HHS Blueprint for Action on Breastfeeding (HHS, 2000b), echoes the recommendations of AAP. In addition to its support, however, the Blueprint identifies certain conditions that warrant caution toward breast- feeding, including exposure to contaminants such as DLCs. These cautions are aimed specifically at high-risk circumstances, such as high-level exposures to contaminants that result in poisonings and the consumption of contaminated noncommercial fish and wildlife (HHS, 2000b). Aside from these specific cir- cumstances, the Blueprint supports the goals of Healthy People 2010: Objectives for Improving Health: to increase the proportion of mothers who breastfeed to 75 percent in the early postpartum period, 50 percent at 6 months, and 25 percent at 1 year (HHS, 2000a). Two ongoing, nationwide, cross-sectional surveys, NHANES and the Ross Laboratories Mothers Survey, have contributed to the knowledge of rates and trends in breastfeeding patterns in the United States. Results from NHANES III, covering the years 1988-1994, indicate that 47 percent of newborns were exclu- sively breastfed at 7 days after birth, and that 10 percent were still exclusively breastfed at age 6 months. Nonexclusive breastfeeding was seen in 52 and 22 percent of infants at 7 days and 6 months, respectively. Similarly, the Ross survey, from data collected in 2001, indicated that 46.3 and 17.2 percent of newborns were exclusively breastfed in-hospital (birth) and 6 months respec- tively, whereas nonexclusive breastfeeding was seen in 69.5 and 32.5 percent of infants in-hospital and at 6 months, respectively. An additional and important finding from the Ross survey was that the greatest increase in both initiation and continuation of breastfeeding between 1989 and 1995 occurred among those most at risk for poor nutritional status: women of color receiving benefits from the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), women with low education, first-time mothers, and women employed full time and residing in regions where breastfeeding is not routinely practiced (Ryan, 1997~. Recent U.S. data on DLCs in breast milk are limited. A 1986 study analyzed breast-milk samples for polychlorinated biphenyls (PCBs) and p,p'-dichloro- diphenyldichloroethylene in more than 800 predominantly white, educated women living in North Carolina (mean age of 27 years). The median PCB con- centration in breast milk was 1.8 ppm (fat basis) at birth, declining by almost 50 percent to 1.0 ppm by the infant's first birthday (Rogan et al., 1986~. This study indicated that exposure to DLCs in breastfed infants was many times higher than exposures that would be expected for adults, but that these exposures declined after weaning, most likely as a result of lower DLC intakes from foods and the infants' expanding total body-fat pool. Feeley and Brouwer (2000) reported simi-
130 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY lar observations for PCBs, polychlorinated dibenzo-p-dioxins (PCDDs), and poly- chlorinated dibenzofurans (PCDFs) in U.S. and European cohorts in the early 1990s, and Furst and colleagues (1989) also observed that DLC concentrations were highest during the first weeks of lactation. A nursing infant's exposure to DLCs per unit of body weight will always be greater than his or her mother's exposure because DLCs are concentrated in the breast-milk fat. After weaning, since other environmental exposures are lower than the exposure to DLCs through breast milk, DLC levels in the infant's body fat will decline to adult levels as the adipose tissue pool size increases. A study of congener-specific DLCs in nonpooled breast-milk samples from Belgium showed that the major DLC congeners in these samples were 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD), 1,2,3,7,8-pentachlorodibenzo-p-dioxin (PeCDD), 2,3,4,7,8-pentachlorodibenzo-p-furan (PeCDF), and PCB-126. The mean value of DLCs present in breast-milk samples was 29.4 pg TEQ/g of fat. The estimated average intake by breastfeeding infants was 76 TEQ pa/kg/d, which is 20-fold higher than the World Health Organization's tolerable daily intake value for adults (Focant et al., 2002~. A similar study on Korean women and infants showed that the predominant congeners in breast-milk samples were 2,3,4,7,8-PeCDD and PCB-126, and that the mean daily DLC intake for breast- feeding infants was 85 TEQ pa/kg/d (Yang et al., 2002~. DLC Exposure of Infants Fed Formula versus Breast Milk. In general, infant formulas have been found to contain significantly lower levels of DLCs than breast milk, due largely to the use of plant oils in place of animal fats in the formula (Ramos et al., 1998; Schecter et al., 1994~. Bajanowski and colleagues (2002) compared tissue levels of PCDD/PCDF congeners in both breastfed and formula-fed infants from Germany that died unexpectedly in the years 1991 to 1992 and 1996 to 1997. Among infants of similar age, the mean concentration of DLCs in fatty tissue from breastfed infants was greater than from formula-fed infants, and concentrations were higher in infants breastfed for longer compared with shorter time periods. The average weekly dioxin intake (I-TEQ ng/kg of fat) was 10- to 25-fold higher in breastfed than in formula-fed infants. When cohorts from 1991 to 1992 were compared with those from 1996 to 1997, total DLC levels were lower in the latter, independent of length of breastfeeding, reflecting the trend toward lower DLC exposure seen in the population at large. Altogether, DLC intake exposure to infants through breastfeeding is signifi- cantly greater than through formula feeding. Evidence indicates that, for the general population, the benefits of breastfeeding outweigh existing evidence for the detrimental effects to infants resulting from exposure to DLCs through breast milk (Work Group on Breastfeeding, 1997~. Therefore, recommendations limit- ing breastfeeding as a means of reducing DLC exposure for the general popu- lation would result in unnecessary loss of the physiological advantages, psy- chological health, and societal benefits of breastfeeding. There are, however,
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 131 documented incidences of unintentional high exposure due to unintended indus- trial releases in which breastfeeding may not be advisable; these situations should be addressed apart from recommendations to the general population. Young Children. The main route of DLC intake exposure in growing chil- dren is through consumption of whole milk and other full-fat dairy products. Furthermore, as suggested by the previously discussed intake exposure scenario, consumption of added fats from all animal sources increases exposure to DLCs. Discretionary fat intake, particularly from animal foods, can be reduced in the diets of young children so that DLC intake exposure can be decreased without compromising the overall nutritional quality of the diet. Additionally, using low-fat or skim milk rather than whole milk will reduce DLC exposure without compromising this important source of calcium for chil- dren. (FDA defines milk that contains 1 percent or less fat as low fat, and 2 percent or less fat as reduced fat.) Results from the 1994-1996 CSFII indicate that 47 percent of children ages 1 to 5 years who consumed milk had whole milk, while 37 percent had low-fat milk. The consumption of milk and other dairy products by children offers a lever- age point for reducing dietary exposure to DLCs for two reasons. First, substitu- tion of low-fat and skim milk and other products for whole-milk counterparts carries no risk of compromising the nutritional quality of the diet for children over the age of 2 years. In fact, children would benefit from the standpoint of reduced long-term risk for chronic disease. Second, federal child nutrition pro- grams, including WIC, the National School Lunch and School Breakfast Pro- grams, and the Child and Adult Care Food Program, reach very large numbers of children and account for a large proportion of their milk consumption. Adolescents 12 to 19 Years of Age. Avoidance of dietary exposure to DLCs by girls before they enter their childbearing years is of importance because of future DLC exposure to a developing fetus through the mother's total body burden. However, many popular food choices among adolescents are high in fat, including animal fat. Lin and colleagues (1996,1999) reported that consumption of foods from fast-food restaurants represents approximately one-third of the average adolescent's meals eaten outside the home. French and colleagues (2001) recently described a strong association between high frequency of fast-food con- sumption and greater intakes of total energy and percentage of energy from fat and a lower intake of milk. Whether these trends translate into increased exposure to DLCs from foods has not been determined. Bearing in mind the long half-life of DLCs in humans, and that adolescents will soon be entering their child-bearing years, it may be wise to monitor and evaluate potential DLC exposure of this group through the consumption of foods high in animal fats.
32 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY American Indian, Alaska Native, and Northern Dwelling Fish Eaters. The native peoples of America differ markedly in culture, language, lifestyle, and diet. Some of those who embrace traditional customs and foodways have in common a dependence on hunting, fishing, and gathering for foodstuffs, whereas others embrace agricultural traditions. Importantly, for a subset of native peoples, a major portion of their diet and cultural practices involves fish and marine mammals, and these groups are at particular risk of exposure to DLCs. In addi- tion, for some agriculturally-based American Indian populations, the lands used for crops and grazing have been previously contaminated with herbicides con- taining DLCs (Personal communication, T. Dawes, Inter Tribal Council of Ari- zona, February 19, 2002~. Their exposure is a function of both the quantity of fish, marine mammals, and agricultural products consumed, and the levels of contaminants in the foods. There is limited information on the extent of fish consumption among vari- ous American Indian people. Fitzgerald and colleagues (1995) and Hwang and colleagues (1996) studied exposure to PCBs among the Mohawks in northern New York. The traditional Mohawk fishing grounds are immediately down- stream of three aluminum foundries that have contaminated the water with PCBs, resulting in high levels of these compounds in the fish (Sloan and Jock, 1990~. Gerstenberger and colleagues (1997,2000) studied fish consumption in an Ojibwa population in Wisconsin. Both of these investigations found that, on average, the frequency of fish consumption was not different from the general population, although some individuals within the populations may have been at higher risk than others. Native peoples living in the far north have fewer local sources of food compared with those living in more temperate climates, and a number of tribal populations are dependent upon fish or marine mammals to varying degrees. In the Americas, most studies have been conducted in northern Canada, particularly with the Inuit people. Dewailly and colleagues (1994) compared an Inuit pop- ulation's dietary exposures to DLCs through fish consumption with the expo- sures of a group of Caucasian fishermen on the north shore of the Gulf of St. Lawrence. The Inuit ate many times greater amounts of fish than the Caucasians did, and they had significantly higher (eight to ten times) plasma levels of PCBs (Dewailly et al., 1994~. The predominant source of DLCs in the diets of other Northern Dwellers is sea birds, sea bird eggs, fish, seal, and whale. Muckle and colleagues (2001) report that Inuit women of child-bearing age living in Nunavik (northern Quebec) ate, on average, 3.3 fish meals per week, and that almost 80 percent ate fish at least once per week. In addition, they consumed beluga whale fat, muktuk (0.5 meals/wk), and meat (0.1 meals/wk), as well as seal fat (0.3 meals/wk), meat (0.4 meals/wk) and liver (0.1 meals/wk). These researchers also found that during pregnancy, 42 percent of the women increased their consumption of traditional foods. Ayotte and colleagues (1997) reported
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 133 that Inuit living in Nunavik (Arctic Quebec) had seven times greater mean total DLC TEQs in breast milk than did women living in southern Quebec. Scrudato and colleagues have an ongoing study of the Yupik Eskimos from St. Lawrence Island, Alaska (Carpenter et al., in press). This island is at the Arctic Circle, 40 miles from the Russian coast. The diet of the Yupiks is primarily whale, seal, and walrus, with berries and greens in the summer. Because of the high-fat content of their food and the likelihood that DLCs are present in the fat, this population is at greater than average risk of exposure to DLCs in their food supply. Although specific DLCs were not tested, initial studies of serum in Yupiks have shown a mean total PCB concentration of 7.1 ppb (range 1.2-19.4), higher than the values reported by ATSDR (2000) for unexposed populations (0.9-1.5 ppb). Subsistence Fishers. There are few studies of non-Native subsistence fisher- men, partly because they are difficult to identify and often do not purchase fishing licenses. Frequent consumption of game fish is of concern because many of these fish have particularly high levels of DLCs (Anderson et al., 1998~. Subsistence fishing is a common cultural tradition in many African-American families and Southeast-Asian immigrants. The Executive Director of the Missis- sippi Rural Development Council (Rawls, 2001) stated that many rural African- American individuals eat fish two to three times per week and, in some rural areas, as often as five times per week. Waller and colleagues (1996) investigated Lake Michigan fish consumption by urban, pregnant, and poor African-American women in Chicago. Ninety percent said that they ate fish regularly, and of these, approximately 10 percent ate sport-caught fish on a regular basis. Hutchison and Kraft (1994) documented fishing activities and fish consump- tion among Asian Hmong residents of Green Bay, Wisconsin. They found that 60 percent of Hmong families fished and one-fourth of these families ate fish at least once a week, including fish from locations covered by fishing advisories. Wong (1997) surveyed 228 people fishing in the San Francisco Bay, of which 70 per- cent were non-Caucasian, especially Asian. The average fish consumption by this population was 32 g (1.2 oz) per day, or about four meals per month. Sport Fishers. Because DLCs are lipophilic and bioconcentrate in the food chain, there are significant levels of these compounds in predatory fish that live in contaminated waters (Kuehl et al., 1989; Schmitt et al., l999~. For example, total PCB levels in Chinook salmon from Lake Huron (338 ppb) and Lake Ontario (835 ppb) (Jordon and Feeley, 1999) exceeded the U.S. Environmental Protection Agency guidelines for unlimited consumption of 50 ppb. Frequent consumption of these contaminated fish may result in undesirable exposure levels. A number of studies have been conducted with sport fishers who fished from the Great Lakes. A study of DLCs in breast milk from 242 women who reported
34 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY moderate consumption of Lake Michigan fish over an average span of 16.1 years, compared with 71 women who reported no consumption, found that fish con- sumption correlated with levels of PCBs in maternal serum and breast milk (Schwartz et al., 1983~. Of another 1,820 women who held Great Lakes fishing licenses (ages 18-40 years), 979 reported that they never ate Great Lakes fish and 408 had not eaten them recently (Mendola et al., 1995~. Among consumers, the average consumption in 1990 to 1991 was 4.9 kg, and the estimated mean life- time sport-fish consumption was 57.4 kg. Anderson and colleagues (1998) compared levels of dioxins, furans, and PCBs in 32 sport-fish consumers from three of the Great Lakes. On average, these individuals ate 49 sport-fish meals per year for a mean of 33 years. Dioxin concentrations in the sport-fish consumers were 1.8 times, furans 2.4 times, and PCBs 9.7 times greater than background dioxin concentrations in the general population. Another 100 Great Lakes fish eaters reported an average of 43 sport- fish meals in the past year. Their dioxin TEQs averaged 9.6 (range 3.0-41.2), furan TEQs 7.4 (range 2.7-25.6), and coplanar PCB TEQs 4.3 (range 0.1-66.2) (Falk et al., 1999~. In two Canadian communities, Great Lakes sport-fish con- sumption averaged 21 g/d among consumers (Kearney et al., 1999~. The mea- sured PCB concentrations in these populations were 4.0 ,ug/L in nonfish eaters, and 6.1 ,ug/L in fish eaters. A recent study noted that the total quantity of sport fish consumed by Lake Michigan anglers declined significantly over the period 1979 to 1993 (He et al., 2001), presumably as a result of an increased knowledge of advisories. Exposure to DLCs from fish consumption is not limited to fresh-water fish. Fatty fish in the Baltic Sea, such as salmon and herring, have levels of PCDDs and PCDFs (Rappe et al., 1989) comparable to those of fish from the Great Lakes (Kuehl et al., 1989~. One study of Swedish men who ate significant amounts of Baltic Sea salmon and herring found that fish consumers had on average 8.0 pa/ g of serum lipid TCDD, compared with 1.8 pg/g of serum lipid for nonconsumers (Svensson et al., 1991~. In summary, persons who consume fatty fish are at risk of greater than average exposure to DLCs if it is a major part of their diet and if they fish from contaminated waters. Certain groups of American Indians and Alaska Natives, for whom fish is a central part of their traditional diet, may be at greatest risk from exposure. Sport fishers may also have high DLC exposure, depending on the levels of DLCs in the fish they consume. Factors Affecting Food Consumption Dietary-consumption patterns are influenced by many different factors, from economic to cultural. Between 1970 and 1997, the consumption of animal-based foods, including red meat, fish, poultry, cheese, fats, and oils, increased in the
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 135 United States, whereas the consumption of milk and eggs decreased (Putnam and Allshouse, 1999~. Among the factors identified as contributing to overall changes in food- consumption patterns among the general population were changes in price and disposable income, increased food availability to the poor, more convenience foods and food imports, an increase in eating away from home, changes in adver- tising strategies, and changes in food-enrichment standards and fortification poli- cies. Additionally, changes in sociodemographics have contributed to current food-consumption patterns. In the United States, there are trends toward smaller household size, greater numbers of households with dual incomes, more single- parent heads of household, and an older, more ethnically diverse population (Putnam and Allshouse, 1999~. The availability and content of consumer information also influence food choices. Many messages and media for messages about food choices and health effects exist. Food choices are most likely to be influenced by this type of infor- mation when the content is meaningful, consistent, and delivered through media accessible to the consumer. Consistency refers to consumers learning the same message, associated with different health effects, from different information sources. For example, educational programs may focus on the benefits of low-fat foods for cardiovascular health benefits, as well as on the reduction in DLC exposure. DLC-reducing behavioral advice is more likely to be effective if it supports, and is consistent with, other food choice behaviors that are recommended for other reasons and that are important to the consumer. Food choices are likely to be unaffected if information efforts do not recognize the cultural importance associated with preparing and eating food. Advice about reducing exposure to DLCs by modifying food-consumption practices will be more successful if the behaviors suggested (e.g., food preparation, choice of food types) are compatible with existing social, cultural, and familial norms. Ethnic Variation in Food-Consumption Patterns Consumption of whole milk and full-fat dairy products represents the major source of DLC intake exposure among children in the United States. The U.S. Dietary Guidelines recommend reducing total fat and saturated fat intake for Americans over 2 years of age (USDA/HHS, 2000~. For children over 2 years of age, whole milk may be a major source of fat that can be reduced to decrease DLC exposure and, at the same time, comply with current dietary recommenda- tions. A study of milk consumption patterns of non-Hispanic white, African- American, and Hispanic New York State WIC participants, ages 1-5 years, showed that fewer African-American and white children drank whole milk with increasing educational level of the adult caregiver. However, Hispanic children
136 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY tended to consume whole milk equally, regardless of the caregiver's level of education (Dennison et al., 2001~. The authors suggested that cultural and lan- guage barriers present in the Hispanic population may have contributed to the study findings. Klesges and colleagues (1999) investigated the milk-consumption patterns of young adult Air Force recruits of various ethnicities. Among 17- to 24-year- old male and female recruits, milk consumption was highest among younger non- Hispanic whites, followed by Hispanics, then Asians, and lowest among African Americans. Within each ethnic group, milk consumption was greater among men than women and was negatively associated with age and education level. Changing Dietary Patterns Among Northern Dwellers A change in food-consumption patterns among northern-dwelling Inuit popu- lations away from traditional foods to more marketed foods common to a West- ern diet has become a unique nutritional dilemma. Inuit women traditionally rely on caribou, seal, and narwhal mattak for protein, iron, and zinc. Similar foods, such as liver, mattak, kelp, and fish skin have, in the past, been this population's source of calcium and vitamin A (Kuhnlein, 1991~. As the use of traditional Inuit foods has declined and been substituted with marketed foods, the risk for inad- equate intake of important nutrients has increased in the population. The di- lemma, however, is that the native species consumed by the Inuits are contribu- tors of important nutrients but are also sources of DLCs (Kuhnlein, 1991~. Although the impact of the consumption of these compounds at the levels present in the Inuit food supply is not fully understood, the substitution of the traditional diet with a more Western diet appears to be associated with other nutritional risks. Vegetarian Food Patterns Given the relatively lower contribution to DLC intakes from fruit, vegetables, and grains than from animal products, a lower body burden in vegetarians might be anticipated as compared with carnivores. Welge and colleagues (1993) found no difference in the mean blood levels of several different PCDD/PCDF conge- ners in 24 vegetarians compared with 24 nonvegetarian volunteers. Importantly, however, this study did not distinguish between facto-vegetarian and vegan co- horts and facto-vegetarian cohorts who consumed some foods that contained animal fats. In a group of strict vegetarians, Schecter and Papke (1998) found that their mean blood DLC concentrations were lower than those of the general population. In another study, Noren (1983) analyzed breast milk for pesticides and PCBs from facto-vegetarian and nonvegetarian women who frequently consumed fatty fish from the Baltic sea. The mean concentrations of four of the six compounds
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 137 tested, adjusted for age and parity, were lowest in the milk of the facto-vegetarian mothers compared with that of the nonvegetarian and fish-consuming mothers. These studies point out the contribution of DLCs in animal-based foods to total DLC intake exposures. Furthermore, the long half-life of DLCs in the human body suggests that changes in body burdens that may occur from dietary modifi- cation will likely take many years to realize. Cooking Methods and Preparation It is possible to reduce DLC exposure through the selection, handling, prepa- ration, and processing of foods at the consumer level. These specific practices can form the basis for targeted consumer education. For dairy products, the selection of low-fat or skim milk (for individuals over 2 years of age) and low-fat dairy products in preference to full-fat products is an effective way to preserve the nutritional contribution of these foods yet minimize DLC exposure. This choice is consistent with the trend over the last three decades toward more use of low-fat and skim milks. However, available data show that approximately 47 percent of U.S. children still consume only or mostly whole milk (Enns et al., 1997~. Cheese is also a significant source of animal fat for most of the population. While low-fat versions of cheeses are available, they are not widely consumed. For children under 2 years of age, the committee concluded that concern for their total caloric requirements and their developmental immaturity outweighs risk, as currently known, from exposure to DLCs. The stability of DLCs and their persistence in the lipid fraction of animal foods make this route of exposure a concern. Several studies have shown that trimming prior to or discarding fat after broiling, pan-frying, grilling, roasting, and pressure-cooking meats decreases DLC levels in the foods by 50 percent or more (Petroske et al., 1998; Rose et al., 2001; Schecter et al., 1998~. Rose and colleagues (2001) evaluated the effect of various cooking methods on dioxin and furan levels in cuts of beef from a single cow that had been dosed with five different congeners. Several different cooking methods were used: frying, grill- ing, and barbecuing ground beef; conventional and microwave oven-roasting beef; and open-pan and pressure cooking stew meat. The cooking process, per se, did not significantly change the concentrations of dioxins and furans. However, losses of fat in preparation and cooking did lower the total amount of the various congeners in the finished product. Schecter and colleagues (1998) determined total pg TEQ and TEQ/kg wet weight for DLCs before and after broiling market basket samples of beef, bacon, and catfish. For all of the samples, total pg TEQ was reduced by an average of 50 percent from broiling. However, the concentrations, as measured by wet weight, were more ambiguous. The beef patty showed no change, the bacon sample greatly increased, and the catfish sample modestly decreased in TEQ/kg wet weight in cooked versus raw product (Schecter et al., 1998~.
138 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY A study comparing pan-fried hamburger with raw hamburger showed re- duced DLC levels in the cooked product by up to 50 percent. The loss of DLCs correlated with the loss of lipid collected from the pan juices after cooking (Petroske et al., 1998~. Clearly, food preparation methods that retain, recycle, or add animal fat into products will increase the potential DLC exposure through those products. Examples are gravies made from meat fat and juices and using lard, bacon grease, or butter for frying, sauteing, or adding to mixed dishes. Selecting lean cuts of meat and removing visible fat is clearly important. For fish and poultry, discarding the skin and trimming visible fat will also reduce exposure to DLCs. One study showed that trimming the lateral line and belly flap from fish and cooking skin-off filets decreased PCB levels in the cooked product by 33 percent (Zabik and Zabik, 1999~. The cooking process neither forms nor degrades DLCs, but substantial decreases in the DLC levels of the prepared product can be achieved by cooking methods that result in the loss of fat. For example, one study found that by increasing the surface area and internal tem- perature of restructured carp surimi products, DLCs from the finished product were 55 to 65 percent lower than in the raw product, depending on the surface area of the fillet (Zabik and Zabik, 1999~. With the exception of members of the squash family (which also have a waxy lipophilic coat and can extract vapor phase or particulate-bound DLCs from the air), most contamination of plants by DLCs appears to be a result of surface deposition of contaminated soils or particulates (Lovett et al.,1997; Muller et al., 1994~. One study (Hulster et al., 1994) showed that PCDD/PCDF detected in squashes (zucchini and pumpkin) came from root uptake, apart from the DLCs that adhered to the plant surfaces. An analysis of carrots, lettuce, and peas grown in highly contaminated soils compared with those grown in background soils revealed that DLCs in the whole vegetable grown in the contaminated soil was greatest for carrots, and that more than 75 percent of the contamination was in the peel (Muller et al., 1994~. Likewise, DLC contamination of peas, although far less than in carrots, was concentrated in the pods. This analysis supports the observation that DLC levels in plants are related to the DLC content of the soils in which they are grown, but that peeling carrots and removing the outer pods from peas are effective ways to reduce DLC intakes from these vegetables. Tsutsumi and colleagues (2000) and Hori and colleagues (2001) assessed the effect of washing and boiling on levels of PCDD/PCDFs and dioxin-like PCBs in samples of green vegetables from supermarket and home-garden sources. These reports showed that washing and cooking lowered the levels of detected DLCs, particularly for the home-garden samples. In addition, Hori and colleagues (2001) suggested that washing was most effective for reducing the levels of dioxins and furans in the samples, but was less effective for reducing the levels of dioxin-like PCBs. In summary, food selection, handling, and preparation practices can provide the consumer with some tools to reduce DLC exposure that are essentially cost-
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 139 free and easy to implement. Selecting low-fat products; trimming and discarding visible fat from meat, fish, and poultry; discarding skin from fish and poultry; avoiding practices that add or retain animal fat; and washing vegetables and peeling root and waxy-coated vegetables are recommended. Food and Nutrition Assistance Programs USDA administers several food assistance programs that provide a variety of benefits to target recipients. The four primary programs are the Food Stamp Program, the National School Lunch Program, the School Breakfast Program, and WIC. The nationwide average monthly participation rate for the Food Stamp Program was just over 19 million in fiscal year 2002 (FNS,2003b). More than 28 million children participated in the National School Lunch Program in fiscal year 2002, and more than 8.1 million participated in the School Breakfast Program (FNS, 2003c, 2003d). Statistics for the WIC program show that in fiscal year 2002 nearly 7.5 million people participated in the program (FNS, 2003e). The target recipients for the National School Lunch and School Breakfast Programs and WIC participants are of particular interest to this report. Although surveys of participants in the National School Lunch Program showed that they had greater intakes of energy, protein, and several vitamins and minerals compared with nonparticipants, they also consumed greater amounts of total fat and saturated fat per day than nonparticipants (Devaney et al., 1993~. Data from the School Nutrition Dietary Assessment Study-II show that in the survey years 1998-1999, at least 70 percent of elementary school lunches served met the program standard for calories (one-third of the Recommended Dietary Allowance), but exceeded the program standards for levels of fat and saturated fat. Results also showed that, on average, elementary school lunches provided 33 percent of calories from total fat and 12 percent of calories from saturated fat. At the same time, calories provided from carbohydrate were below the recommended 55 percent (Fox et al., 2001~. WIC providers encourage pregnant women to breastfeed their babies and the program supplies a greater variety and quantity of foods and offers longer partici- pation to breastfeeding participants (FNS, 2003a). Fifty-six percent of WIC par- ticipants initiate breastfeeding in the hospital; however, only 45 percent are still breastfeeding at discharge, and only 16 percent continue breastfeeding to 5 months of age. WIC participants who breastfeed and are most likely to continue breast- feeding are those who receive the special food package for breastfeeding mothers along with information and advice on breastfeeding (IOM, 2002b). WIC participants receive a food package that includes milk, cheese, and eggs. The recent report, Dietary Risk Assessment in the WIC Program (IOM, 2002b), identified adherence to Food Guide Pyramid recommendations, includ- ing consumption of fat, saturated fat, and cholesterol, as an appropriate criterion for reducing dietary risk, recognizing that essentially low-income women and
140 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY children will meet this criterion. However, whether WIC participants are at risk for greater exposure to DLCs than the general population has not been determined. DIETARY GUIDANCE The ultimate goal of dietary guidance and consumer education with regard to DLC exposure is to reduce body burdens across the whole population, and par- ticularly for women prior to pregnancy and breastfeeding. Empowering individu- als, through information, to make dietary changes that reduce DLC exposure is theoretically possible but requires a long-term view. The stability of these com- pounds in the body means that reduction in risk through dietary choice is ineffec- tive in the short run, but it is likely very important over years and decades, particularly if begun in childhood. There are several sources of dietary recommendations and associated educa- tional tools for the general population that are consistent with the reduction of DLC exposure through the reduction of animal fat intake. HHS and USDA jointly developed the Dietary Guidelines for Americans (USDA/HHS, 2000), which provide recommendations to the general population, based on current scientific evidence, about the relationship between diet and risk for chronic disease, and which serve as guides for how a healthy diet may improve nutritional status. By statute, these guidelines now form the basis of federal food, nutrition education, and information programs. The Food Guide Pyramid is designed as an educa- tional tool to translate these recommendations for the public (USDA, 1996~. The U.S. Dietary Guidelines recommend that Americans over the age of 2 years "choose a diet that is low in saturated fat and cholesterol and moderate in total fat," and specifically that individuals should keep their "intake of saturated fat at less than 10 percent of total calories" and should "aim for a total fat intake of no more than 30 percent of total calories" (USDA/HHS,2000~. In this context, the U.S. Dietary Guidelines recommend choosing "vegetable oils rather than solid fats (meat, dairy fats, shortening)" and fat-free or low-fat milks, yogurts, and cheeses, and limiting intakes of high-fat processed meats and organ meats (including liver) and sauces made with cream. Similar to the U.S. Dietary Guidelines, the American Heart Association's (AMA) 2000 dietary guidelines "advocates a population-wide saturated fat intake of less than 10 percent energy, which can be achieved by limiting intake of foods rich in saturated fatty acids (e.g., full-fat dairy products and fatty meats)" (Krause et al., 2000~. Another report, Dietary Reference Intakes for Energy, Carbohy- drate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (IOM, 2002a), is consistent with the Dietary Guidelines and the AHA recommendations in that it identifies the association between the excessive intake of saturated fats with the increased incidence of heart disease and the development of certain cancers. Further, the reports cited above all recommend that individuals strive to
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 141 achieve and maintain a healthy body weight, that is, to avoid gaining excess weight because of the known adverse health consequences of obesity, indepen- dent of the increased body burden of DLCs due to the accumulation of total body fat. The committee did not, however, find evidence to support a reduction in DLC body burden through weight loss. Thus, preventing obesity is preferred as a means of reducing the potential for DLC accumulation. The committee identified two areas in which changes in individual behavior could reduce DLC exposure, but which are not advised because of countervailing risks. These are reductions in fatty-fish consumption below current recommenda- tions of 1 to 2 fish meals per week and in breastfeeding. Each of these are discussed briefly below. Fish A recommendation that emerged very recently and does not appear specifi- cally in the current U.S. Dietary Guidelines recognizes the potential health ben- efits of fish consumption. Unlike the saturated fat in dairy products and meats, the fatty acids in fish have been shown in epidemiological studies to be beneficial to health. The strongest evidence for the importance of omega-3 fatty acids is a reduction in risk of sudden death from cardiac arrest: epidemiological studies show a lower risk with greater fish consumption among both men and women (Daviglus, 1997; Guallar, 1995; Hu et al., 2002~. Two long-chained, highly polyunsaturated fatty acids, docosahexaenoic acid and eicosapentaenoic acid, are thought to be responsible for these health benefits by providing strong antiarrhythmic action on the heart, by serving as precursors to prostaglandins, and by providing anti-inflammatory and anti-thrombotic ac- tions. There is a large body of epidemiological evidence to indicate that con- sumption of omega-3 polyunsaturated fatty acids, which are present in fatty fish, seafood, and certain vegetable oils, is beneficial in preventing heart disease (Conner, 2000; Hu et al., 2002; Siscovick et al., 1995~. Potential adverse effects on immune function and increased clotting time from high intakes of omega-3 fatty acids have also been identified, but there is a need for greater clinical evidence to support the epidemiological findings (IOM, 2002a). Research is un- derway and will help address these issues. Based on current evidence in support of this decrease in risk for heart disease associated with the consumption of omega-3 fatty acids, AHA recommends eat- ing fish two times per week, including fatty fish such as mackerel, lake trout, tuna, and salmon (Krause et al., 2000~. While AHA acknowledges the reported effects of omega-3 fatty-acid supplements in patients with heart disease, it con- cludes that "further studies are needed to establish optimal doses of omega-3 fatty acids...for both primary and secondary prevention of coronary disease..." and that "consumption of 1 fatty fish meal per day (or alternatively, a fish oil supple- ment) could result in an omega-3 fatty acid intake...of approximately 900 ma/
42 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY day, an amount shown to beneficially affect coronary heart disease mortality rates in patients with coronary disease" (Krause et al., 2000~. In addition to heart disease, some studies suggest that there are benefits of omega-3 fatty acids against a variety of health conditions (Conner, 2000~. These include inflammatory diseases such as rheumatoid arthritis (Carder et al., 2002), epileptic seizure (Schlanger et al., 2002), endometrial cancer (Terry et al., 2002), age-related macular degeneration (Seddon et al., 2001), prostate cancer (Terry et al., 2001), and premature birth (Allen and Harris, 2001~. The increasing evidence of health benefits from the consumption of fish argues that efforts to reduce the DLC content in fish should be promoted, rather than restricting intake in order to reduce DLC exposure. One caution regarding fish consumption exists with regard to women who are pregnant or who may become pregnant: some species of fish, particularly long-lived, larger predatory fish may contain high levels of methylmercury, a developmental neurotoxin that may harm a fetus's developing nervous system if ingested at high levels. FDA has issued guidance to women of childbearing age, advising them to limit their consumption of fish to an average of not more than 12 oz/wk of cooked fish (approximately two to four servings). FDA also recommended that pregnant or potentially pregnant women eat a variety of fish; avoid eating shark, sword- fish, king mackerel, and tilefish; and check with their state or local health depart- ment to see if there are special advisories on fish caught from freshwater lakes and streams in their local area (CFSAN, 2001~. The degree of contamination of farmed and wild-caught fish varies geo- graphically and with changes in aquaculture practices (Gerstenberger et al., 1997; He et al., 2001~. However, as sources of fats used in fish feeds change, so do the levels of DLCs that are introduced through this route. Thus, accurate data on regional sources that contribute to geographic variability are needed. Addition- ally, the committee recognizes that there are comigrating contaminants that will be found with DLCs in many types of fish, but an examination of these other contaminants was beyond the committee's charge and were not considered in this report. In summary, fish are important sources of nutrients and potentially beneficial omega-3 fatty acids, and they also have cultural importance in the traditional diets of many population groups. But, unlike meats and poultry, fish cannot easily be trimmed to reduce their fat content. Therefore, fish consumption should be encouraged at currently recommended levels (1 to 2 fish meals per week), except for fish caught where known DLC contamination has occurred and fish adviso- . . nes are In p ace. Breastfeeding As previously mentioned, AAP recommends exclusive breastfeeding until the age of 6 months and continued breastfeeding with supplementary foods until
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 143 at least the age of 1 year. The benefits of breastfeeding have been clearly and positively identified, at the level of the infant, the mother, the family, and society (Work Group on Breastfeeding, 1997), and it would be inappropriate to interfere with the current upward trend in breastfeeding rates in the United States because of as yet only theoretical detrimental effects of the DLC levels currently found in breast milk. It is important to emphasize that changing the mother's diet during pregnancy and lactation will not have an impact on reducing DLC levels in her breast milk, since the accumulated body burden is what determines DLC expo- sure to the fetus in utero and to the nursing infant. Rather, the focus on reducing DLC exposure should begin after weaning and throughout childhood, so that the next generation of women will enter their reproductive years with lower body burdens of DLCs. This is a long-term agenda and emphasizes the value of reduc- ing animal-fat consumption for children, with particular benefits for girls. For children, the general dietary guidance that applies to adults is appropriate for children over the age of 2 years, including the use of low-fat or skim milk rather than whole milk (AAP Committee on Nutrition, 1983; AHA, 1983~. SUMMARY The major sources of DLCs in the food supply are animal fats, although small amounts appear in most foods. Exposure can be reduced through the re- duced intake of animal fat by selecting lean cuts of meat, poultry, and fish; trimming visible fat and removing skin, as appropriate; and by selecting low-fat dairy products. DLCs on fruits and vegetables can be reduced through washing and through peeling root and waxy-coated vegetables. In general, these precautions are consistent with current dietary advice, with two major exceptions. Although breast milk is a relatively concentrated source of DLCs, particularly in first lactations, and contributes significantly to body bur- dens, the evidence of the benefits of human breast milk overall outweighs the potential risks from this source. The focus should be on reducing lifetime body burdens through other means so that future generations of breastfeeding mothers will increasingly transmit lower levels through this important source of infant nutriture. The second exception is the conflict between recommendations to con- sume more fatty fish and the fact that some of the highest concentrations of DLCs in the food supply come from these fish, both farmed and from the sea. Some subsets of the population are likely to be at greater than average risk from this source and should be monitored. The accumulating evidence of health benefit from fatty fish (e.g., mackerel and salmon) suggest that rather than avoiding this food, the focus must again be on reducing the content of DLCs in future stocks.
44 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY REFERENCES AAP (American Academy of Pediatrics) Committee on Nutrition. 1983. Towards a prudent diet for children. Pediatrics 71 :78-80. AHA (American Heart Association). 1983. AHA committee report. Diet in the healthy child. Circu- lation 67:141 lA-1414A. Allen KG, Harris MA. 2001. The role of n-3 fatty acids in gestation and parturition. Exp Biol Med (Maywood) 226:498-506. Anderson HA, Falk C, Hanrahan L, Olson J. Burse VV, Nee&am L, Paschal D, Patterson D Jr, Hill RH Jr, The Great Lakes Consortium. 1998. Profiles of Great Lakes critical pollutants: A senti- nel analysis human blood and urine. Environ Health Perspect 106:279-289. ATSDR (Agency for Toxic Substances and Disease Registry). 2000. Toxicological Profile for Poly- chlorinated Biphenyls (PCBs). Atlanta: ATSDR. Ayotte P. Dewailly E, Ryan JJ, Bruneau S. Lebel G. 1997. PCBs and dioxin-like compounds in plasma of adult Inuit living in Nunavik (Arctic Quebec). Chemosphere 34:1459-1468. Bajanowski T. Furst P. Wilmers K, Beike J. Kohler H. Brinkmann B. 2002. Dioxin in infants An environmental hazard? Int J Legal Med 116:27-32. Bernt KM, Walker WA. 1999. Human milk as a carrier of biochemical messages. Acta Paediatr Suppl 88:27-41. Calder PC, Yaqoob P. Thies F. Wallace FA, Miles EA. 2002. Fatty acids and lymphocyte functions. Br JNutr 87:S31-S48. Carpenter DO, DeCaprio AP, O'Hehir D, Scrudato RJ, Apatiki L, Kava J. Gologergen J. Miller PK, Eckstein L. In press. Organochlorine contaminants in serum of Siberian Yupik people from St. Lawrence Island, Alaska. Submitted to Environ Health Perspec. CFSAN (Center for Food Safety and Applied Nutrition). 2001. An Important Message for Pregnant Women and Women of Childbearing Age Who May Become Pregnant About the Risks of Mer- cury in Fish. Online. U.S. Food and Drug Administration. Available at http://www.cfsan.fda. gov/%20~dms/admehg.html. Accessed June 4, 2003. Conner WE. 2000. Importance of n-3 fatty acids in health and disease. Am J Clin Nutr 71:171S- 175S. Daviglus ML, Stamler J. Greenland P. Dyer AR, Liu K. 1997. Fish consumption and risk of coronary heart disease. What does the evidence show? Eur Heart J 18: 1841-1842. Dennison BA, Rockwell HL, Nichols MJ. 2001. Use of low-fat milk by children in the New York State WIC varies with parental characteristics. JAm Diet Assoc 101:464-466. Devaney B. Gordon A, Burghardt J. 1993. The School Nutrition Dietary Assessment Study: Dietary Intakes of Program Participants and Nonparticipants. Alexandria, VA: Food and Nutrition Service, U.S. Department of Agriculture (USDA). Dewailly E, Ryan JJ, Laliberte C, Bruneau S. Weber JP, Gingras S. Carrier G. 1994. Exposure of remote maritime populations to coplanar PCBs. Environ Health Perspect 102:205-209. Dewey KG. 2003. Is breastfeeding protective against childhood obesity? J Hum Lact 19:9-18. Dewey KG, Peerson JM, Brown KH, Krebs NF, Michaelsen KF, Persson LA, Salmenpera L, White- head RG, Yeung DL. 1995. Growth of breast-fed infants deviates from current reference data: A pooled analysis of US, Canadian and European data sets. Pediatrics 96:495-510. Douglass JS, Murphy MM. 2002. Estimated Exposure to Dioxins in the Food Supply. Paper prepared for the Committee on the Implications of Dioxin in the Food Supply. Institute of Medicine, Washington, DC. Enns CW, Goldman JD, Cook A. 1997. Trends in food and nutrient intakes by adults: NFCS 1977- 78, CSFII 1989-91, and CSFII 1994-95. Fam Econ Nutr Rev 10:2-15. Falk C, Hanrahan L, Anderson HA, Kanarek MS, Draheim L, Nee&am L, Patterson D Jr. 1999. Body burden levels of dioxin, furans, and PCBs among frequent consumers of Great Lakes sport fish. The Great Lakes Consortium. Environ Res 80:Sl9-S25.
HUMAN FOODS AND FOOD-CONSUMPTION PATTERNS 145 Feeley M, Brouwer A. 2000. Health risks to infants from exposure to PCBs, PCDDs and PCDFs. Food Addit Contam 17:325-333. Fitzgerald EF, Hwang SA, Brix KA, Bush B. Cook K, Worswick P. 1995. Fish PCB concentrations and consumption patterns among Mohawk women at Akwesasne. J Expo Anal Environ Epi- demiol 5:1-19. FNS (Food and Nutrition Service). 2003a. Breastieeding Promotion and Support in WIC. Online. USDA. Available at http://www.fns.usda.gov/wic/Breastfeeding/breastfeedingmainpage.htm. Accessed January 10, 2003. FNS. 2003b. Food Stamp Program: Average Monthly Participation (Persons). Online. USDA. Avail- able at http://www.fns.usda.gov/pd/fsfypart.htm. Accessed June 24, 2003. FNS. 2003c. National School Lunch Program: Participation and Lunches Served. Online. USDA. Available at http://www.fns.usda.gov/pd/slsummar.htm. Accessed June 24, 2003. FNS. 2003d. School Breakfast Program Participation and Meals Served. Online. USDA. Available at http://www.fns.uda.gov/pd/sbsummar.htm. Accessed June 24, 2003. FNS. 2003e. WIC Program: Total Participation. Online. USDA. Available at http://www.fns.usda. gov/pd/wifypart.htm. Accessed June 24, 2003. Focant IF, Pirard C, Thielen C, De Pauw E. 2002. Levels and profiles of PCDDs, PCDFs and PCBs in Belgian breast milk. Estimation of infant intake. Chemosphere 48:763-770. Food Surveys Research Group. 1997. Data Tables: Results from USDA's 1994-96 Continuing Sur- vey of Food Intakes by Individuals and 1994-96 Diet and Knowledge Survey. Table Set 10. Online. Agricultural Research Service (ARS), USDA. Available at http://www.barc.usda.gov/ bhnrc/foodsurvey. Accessed December 9, 2002. Fox MK, Crepinsek MK, Conner P. Battaglia M. 2001. School Nutrition Dietary Assessment Study-II Summary of Findings. Report prepared for the USDA. Cambridge, MA: Abt Associates. French SA, Story M, Neumark-Sztainer D, Fulkerson JA, Hannan P. 2001. Fast food restaurant use among adolescents: Associations with nutrient intake, food choices and behavioral and psycho- social variables. Int J Obes Relat Metab Disord 25:1823-1833. Furst P. Kruger C, Meemken HA, Groebel W. 1989. PCDD and PCDF levels in human milk- Dependence on the period of lactation. Chemosphere 18:439-444. Gerstenberger SL, Tavris DR, Hansen LK, Pratt-Shelley J. Dellinger JA. 1997. Concentrations of blood and hair mercury and serum PCBs in an Ojibwa population that consumes Great Lakes region fish. J Toxicol Clin Toxicol 35:377-386. Gerstenberger SL, Dellinger JA, Hansen LG. 2000. Concentrations and frequencies of polychlori- nated biphenyl congeners in a Native American population that consumes Great Lakes fish. J Toxicol Clin Toxicol 38:729-746. Guallar E, Hennekens CH, Sacks FM, Willett WC, Stampfer MJ. 1995. A prospective study of plasma fish oil levels and incidence of myocardial infarction in U.S. male physicians. J Am Coll Cardiol 25:387-394. He JP, Stein AD, Humphrey HE, Paneth N. Courval JM. 2001. Time trends in sport-caught Great Lakes fish consumption and serum polychlorinated biphenyl levels among Michigan anglers, 1973-1993. Environ Sci Technol 35:435-440. HHS (U.S. Department of Health and Human Services). 2000a. Healthy People 2010: Understand- ing and Improving Health, 2nd ed. Washington, DC: U.S. Government Printing Office. HHS. 2000b. HHS Blueprint for Action on Breastieeding. Washington, DC: Office on Women's Health, HHS. Holladay SD. 1999. Prenatal immunotoxicant exposure and postnatal autoimmune disease. Environ Health Perspect 107:687-691. Hori H. Nakagawa R. Tobiishi K, Iida T. Tsutsumi T. Sasaki K, Toyoda M. 2001. Effects of cooking on concentrations of polychlorinated dibenzo-p-dioxins and related compounds in green leafy vegetables 'Komatsuna'. J Food Hyg Soc Japan 42:339-342.
146 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY Hu FB, Bronner L, Willett WC, Stampfer MJ, Rexrode KM, Albert CM, Hunter D, Manson JE. 2002. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. J Am Med Assoc 287:1815-1821. Huisman M, Koopman-Esseboom C, Fidler V, Hadders-Algra M, van der Paauw CG, Tuinstra LG, Weisglas-Kuperus N. Sauer PJ, Touwen BC, Boersma ER. 1995. Perinatal exposure to poly- chlorinated biphenyls and dioxins and its effect on neonatal neurological development. Early Hum Dev 41:111-127. Hulster A, Muller IF, Marschner H. 1994. Soil-plant transfer of polychlorinated dibenzo-p-dioxins and dibenzofurans to vegetables of the cucumber family (Cucurbitaceae). Environ Sci Technol 28:1110-1115. Hutchison R. Kraft CE. 1994. Hmong fishing activity and fish consumption. J Great Lakes Res 20:471-478. Hwang SA, Fitzgerald EF, Bush B. Cook KSE. 1996. Exposure to PCBs from hazardous waste among Mohawk women and infants at Akwasasne. J Franklin Inst 333A: 17-23. IOM (Institute of Medicine). 2002a. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC: National Academy Press. IOM. 2002b. Dietary Risk Assessment in the WIC Program. Washington, DC: National Academy Press. Jacobson JL, Fein GG, Jacobson SW, Schwartz PM, Dowler JK. 1984. The transfer of polychlori- nated biphenyls (PCBs) and polybrominated biphenyls (PBBs) across the human placenta and into maternal milk. Am J Public Health 74:378-379. Jacobson JL, Jacobson SW. 2002. Association of prenatal exposure to an environmental contaminant with intellectual function in chil&ood. J Toxicol Clin Toxicol 40:467-475. Jacobson SW, Chiodo LM, Jacobson JL. 1999. Breastfeeding effects on intelligence quotient in 4- and 11-year old children. Pediatrics 103:e71. Jordan SA, Feeley MM. 1999. PCB congener patterns in rats consuming diets containing Great Lakes salmon: Analysis of fish, diets, and adipose tissue. Environ Res 80:S207-S212. Kearney JP, Cole DC, Ferron LA, Weber J-P. 1999. Blood PCB, p,p?-DDE, and mirex levels in Great Lakes fish and waterfowl consumers in two Ontario communities. Environ Res 80:S138- S149. Kennedy ET, Bowman SA, Powell R. 1999. Dietary-fat intake in the US population. JAm Coll Nutr 18:207-212. Klesges RC, Harmon-Clayton K, Ward KD, Kaufman EM, Haddock CK, Talcott GW, Lando HA. 1999. Predictors of milk consumption in a population of 17- to 35-year-old military personnel. JAm Diet Assoc 99:821-826. Koopman-Esseboom C, Huisman M, Weislas-Kuperus N. Van der Paauw CG, Tuinstra LGM, Boersma ER, Sauer PJJ. 1994a. PCB and dioxin levels in plasma and human milk of 418 Dutch women and their infants. Predictive value of PCB congener levels in maternal plasma for fetal and infant's exposure to PCB's and dioxins. Chemosphere 28:1721-1732. Koopman-Esseboom C, Morse DC, Weisglas-Kuperus N. Lutkeschipholt IJ, Van der Paauw CG, Tuinstra LG, Brouwer A, Sauer PJ. 1994b. Effects of dioxins and polychlorinated biphenyls on thyroid hormone status of pregnant women and their infants. Pediatr Res 36:468-473. Koopman-Esseboom C, Weisglas-Kuperus N. de Ridder MA, van der Paauw CG, Tuinstra LG, Sauer PJ. 1996. Effects of polychlorinated biphenyl/dioxin exposure and feeding type on in- fants' mental and psychomotor development. Pediatrics 97:700-706.
HUMAN FOODS AND FOOD-CONSUMPTION PA'ITERNS 147 Krauss RM, Eckel RH, Howard B. Appel LJ, Daniels SR, Deckelbam RJ, Erdman JW Jr, Kris- Etherton P. Goldberg IJ, Kotchen TA, Lichtenstein AH, Mitch WE, Mullis R. Robinson K, Wylie-Rosett J. St. Jeor S. Suttie J. Tribble DL, Bazzarre TL. 2000. AHA dietary guidelines: Revision 2000: A statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 102:2284-2299. Kreuzer PE, Csanady GA, Baur C, Kessler W. Papke O. Greim H. Filser JO. 1997. 2,3,7,8-Tetra- chlorodibenzo-p-dioxin (TCDD) and congeners in infants. A toxicokinetic model of human lifetime body burden by TCDD with special emphasis on its uptake by nutrition. Arch Toxicol 71:383-400. Kuehl DW, Butterworth BC, McBride A, Kroner S. Bahnick D. 1989. Contamination of fish by 2,3,7,8-tetrachlorodibenzo-p-dioxin: A survey of fish from major watersheds in the United States. Chemosphere 18: 1997-2014. Kuhnlein HV. 1991. Nutrition of the Inuit: A brief overview. Arctic Med Res S:728-730. Lin B-H, Guthrie J. Blaylock JR. 1996. The Diets of America's Children: Influences of Dining Out, Household Characteristics, and Nutrition Knowledge. Report No. AER-746. Washington, DC: Economic Research Service, USDA. Lin B-H, Guthrie J. Frazao E. 1999. Quality of children's diets at and away from home: 1994-1996. Food Rev 22:2-10. Lovett AA, Foxall CD, Creaser CS, Chewe D. 1997. PCB and PCDD/DF congeners in locally grown fruit and vegetable samples in Wales and England. Chemosphere 34:1421-1436. Lucas A, Morley R. Cole TJ, Lister G. Leeson-Payne C. 1992. Breastmilk and subsequent intelli- gence quotient in children born preterm. Lancet 339:261-264. Mendola P. Buck GM, Vena JE, Zielezny M, Sever LE. 1995. Consumption of PCB-contaminated sport fish and risk of spontaneous fetal death. Environ Health Perspect 103:498-502. Muckle G. Ayotte P. Dewailly E, Jacobson SW, Jacobson JL. 2001. Determinants of polychlorinated biphenyls and methylmercury exposure in Inuit women of childbearing age. Environ Health Perspect 109:957-963. Muller JF, Hulster AA, Papke OC, Ball MC, Marschner H. 1994. Transfer of PCDD/PCDF from contaminated soils into carrots, lettuce and peas. Chemosphere 29:2175-2181. NCHS (National Center for Health Statistics). 2003. National Health and Nutrition Examination Survey. Online. Centers for Disease Control and Prevention. Available at http://www.cdc.gov/ nchs/nhanes.htm. Accessed January 30, 2003. Noren K. 1983. Levels of organochlorine contaminants in human milk in relation to the dietary habits of the mothers. Acta Paediatr Scand 72:811-816. Nutrient Data Laboratory. 2002. USDA National Nutrient Database for Standard Reference, Release 15. Online. ARS, USDA. Available at http://www.nal.usda.gov/fnic/foodcomp. Accessed De- cember 9, 2002. Petroske E, Zaylskie RG, Feil VJ. 1998. Reduction in polychlorinated dibenzodioxin and dibenzo- furan residues in hamburger meat during cooking. JAgric Food Chem 46:3280-3284. Playford RJ, Macdonald CE, Johnson WS. 2000. Colostrum and milk-derived peptide growth factors for the treatment of gastrointestinal disorders. Am J Clin Nutr 72:5-14. Putnam JJ, Allshouse JE. 1993. Food Consumption, Prices, and Expenditures, 1970-92. Statistical Bulletin No. 867. Washington, DC: USDA. Putnam JJ, Allshouse JE. 1999. Food Consumption, Prices and Expenditures, 1970-97. Statistical Bulletin No. 965. Washington, DC: USDA. Ramos L, Torre M, Laborda F. Marina ML. 1998. Determination of polychlorinated biphenyls in soybean infant formulas by gas chromatography. J ChromatogrA 823:365-372.
148 DIOXINS AND DIOXIN-LIKE COMPOUNDS IN THE FOOD SUPPLY Rappe C, Bergqvist PA, Kjeller LO. 1989. Levels, trends and patterns of PCDDs and PCDFs in Scandinavian environmental samples. Chemosphere 18:651-658. Rawls O. 2001. Perceptions of fish safety: Voices from the community. In: National Risk Communi- cation Conference. Proceedings Document. Research Triangle Park, NC: RTI International. Pp. II-17-II-l9. Rogan WJ, Gladen BC, McKinney JD, Carreras N. Hardy P. Thullen J. Tingelstad J. Tully M. 1986. Polychlorinated biphenyls (PCBs) and dichlorodiphenyl dichloroethene (DDE) in human milk: Effects of maternal factors and previous lactation. Am J Public Health 76: 172-177. Rose M, Thorpe S. Kelly M, Harrison N. Startin J. 2001. Changes in concentration of five PCDD/F congeners after cooking beef from treated cattle. Chemosphere 43:861-868. Ryan AS. 1997. The resurgence of breastfeeding in the United States. Pediatrics 9:E12. Schecter A, Papke O. 1998. Comparisons of blood dioxin, dibenzofuran, and coplanar PCB levels in strict vegetarians (vegans) and the general United States population. Organohalogen Compd 38: 179-182. Schecter A, Startin J. Wright C, Kelly M, Papke O. Lis A, Ball M, Olson JR. 1994. Congener- specific levels of dioxins and dibenzofurans in U.S. food and estimated daily dioxin toxic equivalent intake. Environ Health Perspect 102:962-966. Schecter A, Dellarco M, Papke O. Olson J. 1998. A comparison of dioxins, dibenzofurans and coplanar PCBs in uncooked and broiled ground beef, catfish and bacon. Chemosphere 37: 1723- 1730. Schlanger S. Shinitzky M, Yam D. 2002. Diet enriched with omega-3 fatty acids alleviates convul- sion symptoms in epilepsy patients. Epilepsia 43:103-104. Schmitt CJ, Zajicek JL, May TW, Cowman DF. 1999. Organochlorine residues and elemental con- taminants in U.S. freshwater fish, 1976-1986: National Contaminant Biomonitoring Program. Rev Environ Contam Toxicol 162:43-104. Schwartz PM, Jacobson SW, Fein G. Jacobson JL, Price HA. 1983. Lake Michigan fish consumption as a source of polychlorinated biphenyls in human cord serum, maternal serum, and milk. Am J Public Health 73:293-296. Seddon JM, Rosner B. Sperduto RD, Yannuzzi L, Haller JA, Blair NP, Willett W. 2001. Dietary fat and risk for advanced age-related macular degeneration. Arch Ophthalmol 119:1191-1199. Siscovick DS, Raghunathan TE, King I, Weinmann S. Wicklund KG, Albright J. Bovbjerg V, Arbogast P. Smith H. Kushi LH, Cobb LA, Copass MK, Psaty BM, Lemaitre R. Retzlaff B. Childs M, Knopp RH. 1995. Dietary intake and cell membrane levels of long-chain n-3 polyun- saturated fatty acids and the risk of primary cardiac arrest. JAm Med Assoc 274:1363-1367 Sloan R. Jock K. 1990. Chemical Contaminants in Fish from the St. Lawrence River Drainage on Lands of the Mohawk Nation at Akwesasne and Near the General Motors Corporation/Central Foundry Division, Massena, NY, Plant. Tech. Doc. 90-1 (BEP). Albany, NY: New York State Department of Environmental Conservation. Smiciklas-Wright H. Mitchell DC, Mickle SJ, Cook AJ, Goldman JD. 2002. Foods Commonly Eaten in the United States: Quantities Consumed per Eating Occasion and in a Day, 1994-1996. Online. USDA. Available at http://www.barc.usda.gov/bhnrc/food survey/Products9496.htrrd. Accessed June 16, 2003. Svensson BG, Nilsson A, Hansson M, Rappe C, Akesson B. Skerfving S. 1991. Exposure to dioxins and dibenzofurans through the consumption of fish. N Engl J Med 324:8-12. Terry P. Lichtenstein P. Feychting M, Ahlbom A, Wolk A. 2001. Fatty fish consumption and risk of prostate cancer. Lancet 357:1764-1766. Terry P. Wolk A, Vainio H. Weiderpass E. 2002. Fatty fish consumption lowers the risk of endome- trial cancer: A nationwide case-control study in Sweden. Cancer Epidemiol Biomarkers Prev 11:143-145. Tsutsumi T. Iida T. Hori T. Yanagi T. Kono Y. Uchibe H. Toyoda M. 2000. Levels of PCDDs, PCDFs and co-PCBs in fresh and cooked leafy vegetables in Japan. Organohalogen Compd 47:296-299.
HUMAN FOODS AND FOOD-CONSUMPTION PATTERNS 149 USDA (U.S. Department of Agriculture). 1996. The Food Guide Pyramid. Home and Garden Bulle- tin No. 252. Washington, DC: USDA. USDA/HHS. 2000. Nutrition and Your Health: Dietary Guidelines forAmericans, 5th ed. Home and Garden Bulletin No. 232. Washington, DC: U.S. Government Printing Office. Waller DP, Presperin C, Drum ML, Negrusz A, Larsen AK, van der Ven H. Hibbard J. 1996. Great Lakes fish as a source of maternal and fetal exposure chlorinated hydrocarbons. Toxicol Ind Health 12:335-345. Welge P. Wittsiepe J. Schrey P. Ewers U. Exner M, Selenka F. 1993. PCDD/F-levels in human blood of vegetarians compared to those of non-vegetarians. Organohalogen Compd 13:13-17. Wold AK, Adlerberth I. 2000. Breast feeding and the intestinal microflora of the infant Implica- tions for protection against infectious diseases. Adv Exp Med Biol 478:77-93. Wong K. 1997. Fishing for Food in San Francisco Bay. Oakland, CA: Save San Francisco Bay Association. Work Group on Breastfeeding. 1997. American Academy of Pediatrics: Breastfeeding and the use of human milk. Pediatrics 100:1035-1039. Yang J. Shin D, Park S. Chang Y. Kim D, Ikonomou MG. 2002. PCDDs, PCDFs, and PCBs concen- trations in breast milk from two areas in Korea: Body burden of mothers and implications for feeding infants. Chemosphere 46:419-428. Zabik ME, Zabik MJ. 1999. Polychlorinated biphenyls, polybrominated biphenyls, and dioxin reduc- tion during processing/cooking food. In: Jackson LS, Knize MG, Morgan JN, eds. Impact of Processing on Food Safety. New York: Kluwer Academic/Plenum Publishers. Pp. 213-231.