Obtaining food consumption data

In order to obtain estimates of exposure, it is necessary to obtain information about the different foodstuffs consumed, preferably with details about then-packaging. However, data with this extent of detail hardly exist, thus the first step must be to obtain data about the consumption of different foodstuffs. Some consumption surveys are focused on individual consumers, whilst others are directed towards household purchases and others to the whole of the population. Today there are three broad categories for sources of data for food consumption, covering the food supply chain, households and individuals, namely  

Eood balance sheets, whilst being regularly updated, primarily consider raw food commodities and do not reflect the packaging of the foodstuffs consumed. However, they are useful in indicating trends that can be used in any exposure assessment. 

Some countries undertake shopping basket or household budget surveys (HBS), which involve simply recording the contents of a shopping basket for a known family size. However this has many shortcomings. Consider a 

In recall approaches a trained interviewer asks individuals what they ate in the immediate past, typically the preceding 24 or 48 hours. The interview can be face to face or over a telephone. A major disadvantage is that it relies on the memory of the interviewees. If the interview is over the telephone, it also relies on details of the individuals, such as body weight, being accurately 

Food frequency questioimaires determine the frequency with which certain foodstuffs are consumed over a given period. Thus it is necessary to predefine the foodstuffs of interest and these may be targeted to a nutrient(s) or food(s) of specific interest. They are rarely conducted for specific contaminants arising from the packaging of the foodstuffs consumed. It is possible (Parmar et al. 1997) to combine the food frequency quesfioimaire with portion sizes (MAFF for example) in order to obtain an estimate of food intake, particularly where consumption data are limited. 

---

Food consumption surveys are often conducted using the dietary recall or food diary methodologies. In the dietary recall technique, either direct face-to-face interviews or telephone interviews may be used to obtain food consumption data. Consumption amounts frequently are estimated using standard equipment, pictures, and diagrams to aid participants in estimating the amount of food consumed. Data from such recall surveys may be used for both chronic and acute dietary risk assessment The food consumption surveys on which the US dietary risk assessment models are based are of the dietary recall type [14]. 

For processing, the data on energy consumption were collected, within the trade, it was travel distance, information on cargo and storage time of various foods. All data was obtained primarily from farmers, processors and traders, absent sufficient data, it was supplemented by data from available databases, especially the Ecoinvent database. 

There is also effort to develop practical ways of measuring intake of additives. The direct measurement of intake continues to be problematical. Duplicate diet studies require a lot of resources, and there are not enough analytical methods available to test for all the additives of interest. Total diet studies, also known as market basket studies, provide a very general picture which can be a useful start to more detailed work on intake, but they too suffer from the same shortage of analytical methods. Biomarker studies are currently problematical. It is difficult to identify metabolites that are both unique to particular additives and can be readily measured in urine. The estimation of additive intake by calculation is still the preferred method, although it requires a large amount of information on both additive levels in food and much data on food consumption. The latter is difficult to obtain without using a lot of resources - many of us have very varied diets ... 

Both of the above diets supplied almost exactly 5 Cal/g of energy. Hopkins reported the food consumption of experimental rats in terms of calories per 100 g of live weight of the animals, and it is a relatively simple task to work from this data and obtain the actual food consumption of the animals. 

Dietary exposure to pesticides (or to xenobiotics in general) is determined by calculating the product of the amount of chemical in or on the food and the total quantity of food consumed. The quantity of chemical potentially consumed in foods can be estimated from data obtained from residue field trials, metabolism studies, and/or monitoring data. Information from these sources is then analyzed with one of several available models containing food consumption factors from surveys conducted by the United States Department of Agriculture (USDA). For calculation of... 

Example L Kono (2004) conducted an ecological study in Japan that compared consumption of selected nutrients with annual colon cancer incidence and mortality across time. Annual per capita consumption data of selected nutrients and foods was obtained from nationwide surveys, which were conducted from 1950 to 2000 and contained data from randomly selected houses across the nation. The food group identified as meat contained poultry and fish. Consumption of red meat was not available for the early years. Annual age-adjusted rates of cancer incidence (from cancer registries) were available for the period 1975 to 1997, and age-adjusted mortality rates were available for the period 1950 and 2000. Means of height and... 

Information on annual per capita food use In the U.S. Is obtained from published (17. 18) and unpublished data from the Economic Research Service of USDA. Quantities are measured of approximately 350 foods that "disappear" Into the food distribution system. The amounts are derived by deducting data on exports, military use, year-end Inventories, and nonfood use from data on production, Imports, and beglnnlng-of-the year Inventories. Because of the complexity of the food distribution system, use of each food Is not measured at the same point In the system. Some foods are In a raw or primary state while others are retail products when their use Is measured. Subsequent losses that occur In processing, marketing, and home use are not taken into consideration. Food supply data used In the preparation of this report reflect, for the first time, revised population estimates for 1970-1981 based on the 1980 Census and revised estimates for fluid milk and cream use from 1909-1981. Dairy product consumption data for 1981 In this paper are preliminary. 

Dietary exposure is calculated as the very simple product of the amount of food consumed and the magnitude of the residue on the food. Unlike the food consumption value, which often is available from only one source (a government survey), the value of the residue portion of the equation may be provided from a wide variety of sources. These values range from the tolerance or MRL value, to data obtained from residue monitoring programs. The use to which various residue data may be put is often dictated by specific science policies within a given coimtry. 

It should be possible to derive a cumulative exposure curve for the whole of the population and, ideally, for consumers only. It is desirable to obtain estimates of exposure for any selected sub-set of the population, to ensure adequate protection of vulnerable groups. Depending upon the quality of the food consumption survey and associated data this may or may not be easy to achieve. If non-consumers are included in an exposure estimate, as is the case in typical per-capita estimates, then the estimate of exposure will be below the actual exposure and the amount of under-estimating will depend upon the ratio of non-consumers to consumers. 

Dietary nuclide intakes are estimated by several methods (WHO, 1983), e.g., selective studies of individual foods, market basket studies, model dish studies and duplicate portion studies. The first procedure involves the estimation of mean dietary nuclide intakes by collecting staple foods, which are consumed by the subject, and then chemically analyzing them. In market basket studies, individual or composite foods obtained from food stores in the area are analyzed. Data on the food consumption rates of the average person then take on an especially important role. Model dish studies involve the preparation of typical dishes based on both food and dish consumption data and analysis of each dish. Duplicate portion studies offer the greatest degree of reality compared to the other methods. At a minimum, all meals consumed by an individual during one day are chemically analyzed. After the accident in the Ukraine, data on the dietary intakes of Cs, °Sr and transuranium have come from analyses of staple individual foodstuffs. Total diet studies for Ukrainians are scarce. 

Figure 1 Trends in energy consumption by carbohydrate food group as a percentage of total carbohydrate in developed and developing countries, obtained from food balance data in 1994. Data obtained from FAO/WHO (1998). Carbohydrates in human nutrition. Report of a joint FAO/WHO expert consultation. FAO Food and Nutrition Papers 19 66 1-140.

In the United States, the threshold mercury concentration for commercial sale of fish is determined by the Food and Drag Administration, whereas consumption advice for recreational (noncommercial) fish is developed by individual states and tribes. Mercury data collected for development of fish-consumption advisories are typically from analyses of filets (axial muscle tissue, with or without skin) for total mercury, with concentrations expressed on a wet-weight basis. Analysis of filets for total mercury yields a valid estimate of MeHg concentration (Grieb et al. 1990 Bloom 1992), whether the analyzed sample consists of a large filet or a small mass of tissue obtained with a biopsy needle (Cizdziel et al. 2002 Baker et al. 2004). 

According to the FDA Total Diet Study, in which dietary sodium and potassium intakes of three age groups of Americans were estimated from 1977 through 1980, cow s milk supplied 32 to 39% of the total sodium intake for infants, whereas the percentage for toddlers was much lower, 12 to 14% (Shank et al. 1982). Dairy products contributed about 10% of the sodium in the total diet consumed by adults. Data obtained from the National Health and Nutrition Examination Survey (HANES II) 1976-1980 (Carroll et al. 1983) reveal that the median daily consumption of sodium found naturally in food and added during... 

---