Market Baskets

We generally collect TDS “market baskets” (abbreviated “MB”) four times each year, once in each of four regions of the country (North Central, West, South, and Northeast). Each MB collects the specific foods to be analyzed. To the extent possible, the MBs are identical from region to region. Each region consists of three cities that differ every year, in which the foods are collected from supermarkets, grocery stores, and fast-food restaurants. 

The foods collected for the TDS market baskets represent the major components of the average diet of the U.S. population, from infants to adults, based on results of national food- consumption surveys. For the first TDS update shown on this website, in which we had analyzed market baskets collected between 1991 - 2002, we had based the list of foods to be collected on the USDA 1987-88 Nationwide Food Consumption Survey (USDA, 1990).

In the 2003 food-list update, which we began applying to market-basket collections in 2003, we based the food list on the USDA's 1994-96, 1998 Continuing Survey of Food Intakes by Individuals (CSFII) (USDA, 2000). The survey included detailed information on the types and amounts of food consumed by the survey participants, who collectively reported on more than 5,000 different foods (USDA, 2000).

We made this large number of foods more manageable by aggregating the foods into groups according to similarity (for example, a given group might include the various kinds of fish included in the survey). For each group, we then assigned the most-consumed food within that group as the representative of all the foods in the group, and designated it as the TDS food; in other words, each of the 280 foods on the TDS Food List may actually represent several foods. 

Since 2003, we made minor changes to the food list as needed in order to continue making sure the average diet of the U.S. population was represented. 

After we collect the foods, we ship them to a central FDA laboratory for preparation and analysis. For each kind of food, we combine samples from each of the three separate cities of the region to form one composite sample of that particular food.

For each TDS food, the samples from each of the three cities in a region are formed into a composite.

Numerous FDA units participate in the operation of the TDS. The Center for Food Safety and Applied Nutrition (College Park, MD) provides overall management of the program and summarizes the laboratory results. FDA District Offices from across the U.S., under the Office of Regulatory Affairs, collect the foods and ship them to the FDA Kansas City District Laboratory (Lenexa, KS), which prepares and analyzes them for toxic elements, nutrient elements, and pesticides. Portions of the prepared composites from two market baskets per year are sent to FDA's Winchester Engineering and Analytical Center (Winchester, MA) for radionuclide analyses. After we have reviewed and finalized the results of the analyses, we post them on this website.

Where samples are prepared and analyzed. 

Each of the approximately 10-year iterations of the TDS is based not only on an updated food list, but also on updated consumption amounts. To estimate dietary exposure to each TDS analyte, we: 

• determine the average level of that analyte in each TDS food.
• estimate the average amount the U.S. population (or age / gender subpopulation) consumes of each TDS food.
• for each TDS food, multiply the average level of that analyte in that food by the average amount the U.S. population (or age / gender subpopulation) consumes of that food.
• combine the results to arrive at the estimated total dietary exposure to the analyte among the U.S. population (or age / gender subpopulation).

Age and gender groups for whom we have calculated TDS consumption amounts

To calculate average per-capita daily consumption amounts for a given food on the TDS Food List, we first go back to the group of foods from the surveys; that is, the multiple foods that the TDS food represents. We calculate per-capita consumption amounts for each of the foods within that group and combine them into a single figure. We use that single figure as the per-capita consumption amount for that TDS food. (See graphic below.)

You can use a mapping file we created to map any of the more-than 7,900 food codes in the 2003-2014 National Health and Nutrition Examination Survey (NHANES), What We Eat In America (WWEIA) to the TDS Foods (i.e., the TDS Foods List) we analyzed in 2014, to find their representative chemical concentrations. To request the file, contact [email protected]. As noted earlier, each of the TDS Foods represents a large group of similar foods. The concentrations of chemicals in each of the TDS Foods applies to all of the foods in the group the TDS Food represents (a limitation that should be noted when reporting chemical concentrations estimated via the TDS and the TDS mapping file). Data on the TDS Foods’ chemical concentrations can be downloaded  below under Analytical Results. 

TDS uses various methods to analyze foods and beverages for analytes in the following categories:

• toxic and nutrient chemical elements
• pesticides
• industrial chemicals
• radionuclides

We (FDA) update the methods for the analyses as the science and laboratory equipment evolve, and the methods undergo evaluation and testing before being implemented. Analytical Methods (PDF, 106 KB) describes the methods used to detect and measure each analyte. In 2014, we updated the methods of measuring elements in foods. The newer methods can detect and differentiate the elements at lower levels.  If you are comparing TDS elements concentrations in foods over time, be aware that an apparent increase in positive findings starting in 2014 (or in 2016, for iodine) might be due to the improved ability to detect them. For a detailed explanation of these changes, please read 2014 Elements Methods Changes (PDF, 100 KB). 

All TDS foods are analyzed for elements, pesticides, industrial chemicals, and radionuclides. Certain foods are analyzed for mercury. Refer to the food/analyte matrix to determine which TDS foods are analyzed for each analyte group.

TDS foods analyzed for elements, pesticides, and industrial chemicals include results for each of the four market basket (MB) collections that occur each year; radionuclides typically are measured in only one MB each year. Therefore, results for elements, pesticides, and industrial chemicals have four sets of results per analyte, per year; whereas radionuclides typically have one set of results per analyte, per year.

As noted in the introduction, results of the TDS, from 1991 to the present, are available in electronic form on this page, and results prior to 1991 may be found in the publications listed on the publications page of this website.

TDS results from 1991 – 2017 are in the table below and include separate files for elements, radionuclides, and pesticides and industrial chemicals. The data for each analyte group are separated into two zip files for results from 1991 through 2002 and for results from 2003 through 2017. Results prior to 1991 may be found in publications some of which can be found on FDA Total Diet Study Results.

How to access and understand the files below: Each file is a compressed, tab-delimited text file that can be downloaded, decompressed, and imported into a database or spreadsheet. The data in the files are presented in columns. To understand what each column represents, readers must first access the Individual Year Analytical Results Column Key (PDF, 105 KB).

Notes:

1. MB 2017-04: In anticipation of the new sampling plan that began in FY2018, the last market basket of 2017 (2017-04) represents a test run of the new sampling procedure. This new sampling procedure categorizes TDS foods as either regional or national. The test used the TDS Regional Food List which is comprised of foods that typically vary based on where they are purchased e.g. milk, bread, meat, produce. This regional list also includes some new foods like tilapia, walnuts, blueberries, mozzarella cheese, and many more. The data presented consists of results for 86 regional foods. Market baskets 2017-01 through 2017-03 present data for 266 foods collected using sampling procedure described above under Study Design.
2. Nickel: Because of a data quality concern, the 2014, 2015, and 2016 analytical results for nickel are not reported in the individual year analytical results text files of TDS element analyses. In 2017 our laboratory studies and data analysis revealed that the equipment we used to grind foods to prepare samples could have leached small amounts of nickel into the samples. We subsequently have taken corrective actions including using different equipment and changing the sample-preparation procedures and we are continuing to monitor to ensure the effectiveness of these changes.
3. Changes in methods of analysis: In 2014, we updated the methods of measuring elements in food. Compared with previous methods, the newer methods can detect and differentiate the elements at lower levels. If you are comparing TDS elements data in foods over time, be aware that an apparent increase in positive findings starting in 2014 might be due to the improved ability to detect them. Please see Analytical Methods above for more information.
4. Pesticides: Some individual pesticide results use number codes to represent the extraction and determination methods used. The codes appear as three- and two-digit numbers, respectively.  See Pesticide Extraction and Determination Codes (PDF, 491KB) for a key showing which method each of these number codes represents.
5. Radionuclides: In the table above, the years listed for radionuclide results (2005, 2004, and 2003) are followed by an additional number (either 1, 2, 3, or 4). The additional number indicates which of the four market baskets collected for that year included radionuclide measurements; for example, 2005-2 means that the food samples tested for radionuclides in 2005 came from the second market basket collected that year.
6. Benzene: View the results on benzene with caution. An FDA evaluation has determined that the TDS method used in the Kansas City District Office laboratory to measure benzene produces unreliable results for benzene in some foods. Based on this evaluation, FDA scientists recommend that benzene data be viewed with great caution while FDA considers removing these data from the TDS website. There is no evidence of problems with other TDS data. See Questions and Answers on the Occurrence of Benzene in Soft Drinks and Other Beverages for more information.