Shared Left Border
    Main Entities
    Estimating Values
     -  Beef and veal
     -  Pork
         -  USDA regression
         -  Danish Pork 2015
     -  Lamb
     -  Fish
     -  Dairy Products
    Recipe Calculation
    Data Presentation
    Data Interchange
    Site Overview
    DFI Home
    SciName Finder™

Estimating Component Values - Pork

 Updated 2020-07-19

 USDA regression analysis of new pork data

It has always been a challenge to have enough nutrient values to make the nutrient composition profile of a certain food complete. For meat and meat products, in this case pork, this is no exception. Pork exists in a large number of different cuts and quality grades, and this makes the life of a food composition compiler difficult.

In the preface of Agriculture Handbook No. 8-10: Pork Products - Raw - Procced - Prepared, Anderson et al (1992) wrote:

"Data are limited for certain nutrients in pork products for a number of cuts for which presenting values is desirable. For fresh raw pork items and unheated cured items, such values are calculated based on known content of the nutrient in the lipid (fatty acids); total solids (vitamin A and cholesterol); moisture-free, fat-free solids (minerals); or moisture-free, fat-free, ash-free solids (water-soluble vitamins) of a similar for of the foods".

These sentenses are interesting as they indicate some methods of estimatiting component values from other component values.

Due to the need for new data on pork products, USDA Nutrient Data Laboratory, in collaboration with scientists at selected universities and the National Pork Board, has conducted studies to determine the nutrient composition, both raw and cooked, of nine (9) fresh pork cuts (shoulder blade steak, tenderloin roast, top loin chop, top loin roast, sirloin roast, loin chop, rib chop, country style ribs, and spare ribs) and fresh ground pork at various fat levels.
The USDA National Nutrient Database for Standard Reference (SR) has been updated in 2007 and in 2009 with these new data.

Ground Pork Data

Williams et al (2009) describe the collaborative study conducted to determine the mathematical relationship between individual nutrients and fat content of raw ground pork using mixed model regression analysis.
The data have been published in tables that provide nutrient profiles for raw ground pork from 4-28% fat, in increments of 1% fat, as determined by regression equations.
Using reverse engineering on these raw ground pork data, it is possible to deduct the regression equations. The results are listed below.

Regression equations based on total lipids content for ground pork

The USDA equations are based on the content of total lipids in ground pork. Based on the data listed by Williams et al (2009), the following regression equations can be deducted:

  Component Unit Regression equation
  Moisture [g/100 g] =   76.6 - 0.7461 [g moisture/g total lipids] * total lipids [g/100 g]
  Cholesterol [mg/100 g] =   56.35 + 0.7052 [mg cholesterol/g total lipids] * total lipids [g/100 g]
  Protein [g/100 g] =   22.13 - 0.2593 [g protein/g total lipids] * total lipids [g/100 g]
  Ash [g/100 g] =   1.124 - 0.0133 [g ash/g total lipids] * total lipids [g/100 g]
  Thiamin [mg/100 g] =   0.4413 - 0.006817 [mg thiamin/g total lipids] * total lipids [g/100 g]
  Riboflavin [mg/100 g] =   0.3773 - 0.00244 [mg riboflavin/g total lipids] * total lipids [g/100 g]
  Niacin [mg/100 g] =   8.413 - 0.1248 [mg niacin/g total lipids] * total lipids [g/100 g]
  Pantothenic acid [mg/100 g] =   0.6482 - 0.0006057 [mg pantothenic acid/g total lipids] * total lipids [g/100 g]
  Vitamin B6 [mg/100 g] =   0.7063 - 0.009685 [mg vitamin B6/g total lipids] * total lipids [g/100 g]
  Vitamin B12 [ug/100 g] =   0.6089 + 0.007516 u[g vitamin B12/g total lipids] * total lipids [g/100 g]
  Sodium [mg/100 g] =   66.93 + 0.06976 [mg sodium/g total lipids] * total lipids [g/100 g]
  Potassium [mg/100 g] =   331.6 - 5.47 [mg potassium/g total lipids] * total lipids [g/100 g]
  Calcium [mg/100 g] =   15.28 + 0.01116 [mg calcium/g total lipids] * total lipids [g/100 g]
  Magnesium [mg/100 g] =   20.1 - 0.2615 [mg magnesium/g total lipids] * total lipids [g/100 g]
  Phosphorous [mg/100 g] =   199.6 - 2.401 [mg phosphorous/g total lipids] * total lipids [g/100 g]
  Iron [mg/100 g] =   0.8504 + 0.001814 [mg iron/g total lipids] * total lipids [g/100 g]
  Zinc [mg/100 g] =   1.934 - 0.001706 [mg zinc/g total lipids] * total lipids [g/100 g]
  Selenium [ug/100 g] =   36.26 - 0.3815 [ug selenium/g total lipids] * total lipids [g/100 g]


Regression equations based on moisture content for ground pork

As the moísture content and total lipd content has a linear relationship, the regression equations in the following are based on the moisture content, the easiest component to determine:

  Component Unit Regression equation
  Total lipids [g/100 g] =   102.7 - 1.34 [g total lipid/g moisture] * moisture [g/100 g]
  Cholesterol [mg/100 g] =   128.8 - 0.9452 [mg cholesterol/g moisture] * moisture [g/100 g]
  Protein [g/100 g] = - 4.488 + 0.3475 [g protein/g moisture] * moisture [g/100 g]
  Ash [g/100 g] = - 0.2413 + 0.01782 [g ash/g moisture] * moisture [g/100 g]
  Thiamin [mg/100 g] = - 0.2586 + 0.009137 [mg thiamin/g moisture] * moisture [g/100 g]
  Riboflavin [mg/100 g] =   0.1268 + 0.00327 [mg riboflavin/g moisture] * moisture [g/100 g]
  Niacin [mg/100 g] = - 4.4 + 0.1673 [mg niacin/g moisture] * moisture [g/100 g]
  Pantothenic acid [mg/100 g] =   0.586 - 0.0008119 [mg pantothenic acid/g moisture] * moisture [g/100 g]
  Vitamin B6 [mg/100 g] = - 0.2881 + 0.01298 [mg vitamin B6/g moisture] * moisture [g/100 g]
  Vitamin B12 [ug/100 g] =   1.381 - 0.01007 [ug vitamin B12/g moisture] * moisture [g/100 g]
  Sodium [mg/100 g] =   74.1 - 0.0935 [mg sodium/g moisture] * moisture [g/100 g]
  Potassium [mg/100 g] = - 230 + 7.331 [mg potassium/g moisture] * moisture [g/100 g]
  Calcium [mg/100 g] =   16.43 - 0.01496 [mg calcium/g moisture] * moisture [g/100 g]
  Magnesium [mg/100 g] =   6.746 - 0.3505 [mg magnesium/g moisture] * moisture [g/100 g]
  Phosphorous [mg/100 g] = - 46.9 + 3.218 [mg phosphorous/g moisture] * moisture [g/100 g]
  Iron [mg/100 g] =   1.037 - 0.002431 [mg iron/g moisture] * moisture [g/100 g]
  Zinc [mg/100 g] =   1.759 + 0.002287 [mg zinc/g moisture] * moisture [g/100 g]
  Selenium [ug/100 g] = - 2.909 + 0.5114 [ug selenium/g moisture] * moisture [g/100 g]


Fresh Pork Data

Patterson et al (2009) describe the study to determine the nutrient composition of nine (9) fresh pork cuts. Data for the nine cuts are also published, and based on the data, it is possible to establish the regression equations for each component in raw pork cuts.

Regression equations based on total lipids content


Regression equations based on moisture content



  • Anderson B, Dickey L. E., Hoke I. M. (1992):
    Composition of Foods: Pork Products - Raw - Processed - Prepared.
    Agriculture Handbook Number 8-10.
    Human Nutrition Information Service,
    United States Department of Agriculture, revised December 1992.
  • Williams J., Patterson K., Howe J., Snyder C., Boillot K., Lofgren P., Thompson L., Luna A., Nalawade A., Douglass L., Holden J. (2009):
    Nutrient Composition in Ground Pork using Regression Techniques.
    Institute of Food Technologists (IFT), June 6-10, 2009, Anaheim, California.
  • Patterson K. Y., Trainer D., Holden J. M., Howe J. C., Buege  D. R., Williams J. R., Snyder C., Boillot K, Lofgren P., Thompson L., Luna A., Doughlas L. W. (2009):
    USDA Nutrient Data Set for Fresh Pork (from SR), Release 2.0.
    Nutrient Data Laboratory (NDL), Agricultural Research Service, U.S. Department of Agriculture, October 2009.


The French Ciqual 2020 table is online

With 3,185 foods and 67 constituents, the 2020 version of the Ciqual table is available. See the ANSES-CIQUAL website.
First Anniversary release of USDA ARS FoodData Central

For more information, see USDA National Agricultural Library's Food and Nutrition Updates - May 2020.
Report Report on nutrient analysis of key cuts of British pork

Analytical survey on key cuts of pork sold in the UK commissioned by Agriculture and Horticultural Development Board (AHDB) and Public Health England (PHE). See Quadram Institute Bioscience’s Food Databanks National Capability website.
NEVO-Online 2019

New edition of Dutch food composition database NEVO-online: compositional data on more than 2150 food items - 2019. See NEVO website.