William Harris, M.D. -- The Scientific Basis of Vegetarianism
Home | On Going Vegan | Questionnaire | Menu for starting out | Exercise | Water & Salt | Supplements
About Dr. Harris | Visitors | Labels | Pesticides | HippoSuction | Testimony | Computercide | F.L.A.K.E.
The Scientific Basis | Breast Cancer | Vegan Political Action | Health Discussion Board
 

Download the major source of food
composition data in the United States.

The USDA Nutrient Database for Standard Reference, Release 13 was available on CD-ROM in 2000 from the Government Printing Office for $17.00 (including shipping and handling). Orders by telephone, 202-512-1800; FAX, 202-512-2250; or on the internet, http://www.access.gpo.gov/su_docs/sale.html. Stock Number 001-000-04679-9.

Click here to download zipped Excel file abbrev26.zip.

This 19.1 MB Microsoft Excel Spreadsheet "ABBREV26.xls" was assembled from USDA SR13. The zipped version is "ABBREV26.zip" (5.63) MB and requires about 20 minutes to download on a 56K modem. After that it can be unzipped and operated by Microsoft Excel or Corel Quattro Pro.

ABBREV26.xls contains data on 6,210 food items in 171 columns including macro nutrients (carbohydrate, fat, and protein), amino acids, carotenoids, fatty acids, isoflavones, minerals, phytosterols, vitamins, and miscellany. Column headers in regular type indicate nutrient values per 100 grams of food. Headers in italics indicate nutrient values per 100 Calories of food.

Abbrev26.xls consists of multiple sheets: "Database" (the main data sheet), "About", which gives an overview of the entire file, "NUT_DATA" (source of most of the of the USDA data that was unscrambled), "NUT_DEF" (contains nutrient numbers and descriptions), "Sugar", "Trans Fats" (which can not be related to the Nutrient Database Numbers in the "Database" sheet but is included for interest), and "CFB" which gives values for Coumestrol, Formononetin, and Biochanin A. Many of highest CFB values, extracted from Acrobat file "Cbf_tbl.pdf" do not correspond to NDB Nos in "Database," so are not included there.

Sheets added on 8/25/07 include a vegan database of 1556 zero cholesterol (i.e. "Vegan") minimally processed foods, a self calculating Vegan Diet and Exercise Questionaire, a health diary frame, a laboratory report frame, and a diagnostics imaging frame.

Sorting:

To sort the entire "Database" sheet on any specific nutrient put the mouse pointer in cell A8 and hold down Cntrl-Shift-End to "select all". Go next to Data/Sort, select "my list has header row", and "sort by descending". Then left-click the small down carat and pick from the choices in the sort box the desired column (e.g. Calcium/wt or Calcium/Cal) and click "OK". Excel will sort the file in a few seconds and Ctrl-Home brings the pointer back to the top of the food list in column B. Hold down the right arrow key until the appropriate nutrient column shows the descending values.

Other Uses:

It is advisable not to save "ABBREV26.xls" unless it is renamed since any changes made will disrupt the original data. If a sort has been performed it can be undone by Edit/Undo Sort or Ctrl-Z. No provision is made for individual nutritional analysis, however it could be done by copying the top six header rows into a new spreadsheet and then selecting all of the foods eaten in a given period of time and copying those rows into the new spreadsheet. Standard spreadsheet methods could then be used to adjust the food quantities and summarize the nutrient intake.

Construction details:

The original USDA file "Abbrev.dbf" contained 36 columns of macro nutrient, vitamin, and mineral data. Using Adobe Acrobat, Microsoft Access, and Microsoft Excel, the USDA files Cbf_tbl.pdf, Nut_data.dbf, Nutr_def.dbf, Car_tbl.txt, Isfl_tbl.txt, Isfl_dsc.txt, Trns_dat.txt, Trns_doc.txt, Vitk_dat.txt, and Vitk_doc.txt were queried for the remaining data which was then inserted into the resulting 171 column "ABBREV21.xls". Care was taken to insure that the Nutrient Data Base Numbers (Column A: NDB_NO) were properly aligned. Formulas and error messages were removed and replaced by values using Corel Quattro Pro.

The columns containing nutrient/100 Calorie data were derived by dividing the USDA nutrient/100 gm columns by the Calorie/100 gm column (column D) and multiplying by 100.

Percent of Calorie from Carbohydrate (CHO), Fat, and Protein columns used similar methods using the factors 4, 9, and 4 as Calories per gram, respectively. The resulting CHO data sometimes exceeded 100% so the specific numbers were adjusted downward to 100%.

Fatty acids:

The USDA includes (incomplete) data on 25 fatty acids using their biochemical definitions (e.g 18:3 = 18 carbon atoms and 3 double bonds), systematic name (e.g octadecatrienoic), and common name of most typical isomer (e.g Alpha-linolenic or ALA). Only two of them are nutritionally essential, 18:2 LA and 18:3 ALA.

RDAs:

Historical perspective:

1946 Minimum Daily Requirement (MDR) were the standard

Recommended Dietary Allowances (RDA) came next. RDAs are set 2 standard deviations above the mean nutrient requirement to insure that only 2.5% of the population will fall below nutrient needs, but this insures that 97.4999..% of the population meeting the RDAs will exceed nutrient and Calorie requirements.

In the '70s the Food and Drug Administration (FDA) came out with the U.S. RDAs which are different from the RDAs of the Food and Nutrition Board (FNB) of the National Research Council (NRC), of the National Academy of Sciences (NAS) which (parenthetically) is not a branch of the U.S. Government.

Recommended Dietary Intake (RDI) is now in vogue but looming on the horizon are Dietary Reference Intakes (DRIs), an umbrella term for groups of values including RDAs, Adequate intake (AIs), Estimated Average Requirement (EAR), and Upper Intake level (UI).

Someone is having a good deal of fun with the English language

There is an array of differing opinions and units of measurement (should we use International Units? Retinol Equivalents? Alpha-tocopherol equivalents? Niacin equivalents? Micrograms? Milligrams? Grams? calories? Calories? Kilocalories? Kilojoules? Megajoules?). I am aware that including RDAs in this file is tantamount to putting on a shirt that says "Kick me." However I include ranges that seem to be in most ball parks. I take exception only in the following items:

Sodium

1. Sodium RDAs have been set in the range of 1100-3300 mg/day even though it is known that human sodium balance can be maintained on as little as 198 mg/day (1). It is also known that physicians routinely put their patients on 500 mg sodium/day diets without adverse effects, since sodium is a risk factor for hypertensive cardiovascular disease and peripheral edema. I have cut the RDA range to 300-3300 with the observation that a balanced diet will meet the lower figure even without added salt.

Calcium

2. Calcium RDAs have escalated in concert with rising rates of osteoporosis in the U.S. with some authorities suggesting as much as 2500 mg calcium/day. However countries reporting as little as 350 mg calcium/caput/day do not seem to be particularly troubled by osteoporosis. An intake of 800 mg/day would seem prudent, however, with the option of adjusting the number if informed and non-commercial advice warrants it.

Niacin

3. No RDA for Niacin (formerly vitamin B3) is given here, since a milligram can be synthesized from 60 mg of leftover tryptophan and the daily metabolic needs are only around 2.5 mg..

Fatty acids

4. RDAs for the two essential fatty acids linoleic (LA) and alpha linolenic (ALA) are thought to be in the range of 1-2% of Calories (2) which split evenly works out to about 3 grams apiece. The challenge in a grain-based agriculture is keeping the LA:ALA ratio at unity.

USDA Nutrient Tables

USDA SR13 is a successor to USDA Handbook No.8, published in 1963. The first half of that book gave composition of foods in 100 gram edible portions and the second half, composition of foods in 1 pound edible portions. This seemed like a foolish waste of space. Why not have the second half give composition of food in 100 Calorie portions instead of in an even less scientific unit of weight? No doubt weighing food is a simple way to find out how much you've got, but there is no RDA for weight in the diet while there is an RDA for Calories. Most people have caught on to the idea of percent of Calories from fat, protein, and carbohydrate, why not carry that idea to its logical conclusion and give nutrient/Calorie ratios for all the other nutrients, too? The foods with the highest nutrient/Calorie ratios are the ones providing maximum nutrition with the least risk of obesity, arguably the chief dietary problem in the U.S. For the most part those foods also turn out to be plant foods, so vegetarians should find the sorted nutrient/100 Calorie columns of particular interest.

Refs:

1. Guyton AC. Textbook of Medical Physiology. WB Saunders. Philadelphia, 1971. ISBN 0-7216-4392-2

2. Murray RK, Granner DK, Mayes PA, and Rodwell VW.Harper's Biochemistry. Appleton and Lange Norwalk, CT 1990. ISBN 0-8385-3640-9. p575.

Most of the original USDA files are available for download at:

http://www.nal.usda.gov/fnic/foodcomp/Data/SR13/sr13.html

The SR 13 CD-ROM was issued February, 2000. "ABBREV21.xls" was completed on July 10, 2000 by:

William Harris, M.D.

1765 Ala Moana Blvd. #1880

Honolulu, HI 96815

Author of The Scientific Basis of Vegetarianism

http://www.vegsource.com/harris/