Somatic cells: Function and relationship to milk quality

Somatic cell discussions must consider several key points: the source and function of somatic cells, concentration in milk and relationship to milk production, impact on quality of milk and milk products, regulatory limits and comparison of the different methods used to calculate average somatic cell counts.

Dairy cows face continual exposure to bacteria capable of causing many diseases including mastitis. A key element of mastitis control involves bacteria control. This includes maintaining a clean environment to limit cows’ exposure to environmental pathogens, minimizing cow to cow transfer of contagious mastitis organisms at milking time, as well as keeping cows healthy so they can effectively fight off challenges which occur.

Somatic cell discussions, which inevitably are a part of mastitis discussions, must consider several key points. These include the source and function of somatic cells, concentration in milk and relationship to milk production, impact on quality of milk and milk products, regulatory limits and comparison of the different methods used to calculate average somatic cell counts.

Somatic Cell Count-Regulatory Requirements

An important somatic cell concentration reference number from the Pasteurized Milk Ordinance (PMO) (7) is used for regulatory purposes. The PMO currently states the legal maximum somatic cell count for Grade A farm bulk milk is 750,000/ml. Milk buyers can and frequently do offer monetary incentives to producers providing lower cell count milk. The benefits derived from low cell count raw milk make bonus incentives economically viable. The 750,000/ml limit is high compared to many international standards. Much of Europe, New Zealand and Australia have a limit of 400,000/ml and Canada has a limit of 500,000/ml.

The 750,000/ml US limit is based on action taken by the National Conference of Interstate Milk Shippers (NCIMS) which includes producers, scientists, regulatory specialists and processors. Collectively they have established the maximum cell count for grade A raw milk. The maximum was higher in the past and it has been lowered on several occasions. Lowering it further has recently been proposed and debated. To date, those involved in the decision have opted to leave the limit at 750,000/ml.

The US milk quality-monitoring program is based on routine producer sampling and testing. At least one bulk tank sample/month from each Grade A producer must be tested for somatic cell and bacteria counts. When the somatic cell concentration of 3 of the last 5 milk samples exceeds 750,000/ml, producers lose the right to sell milk for Grade A purposes and their license is suspended. To regain the license, actions must be taken to produce milk with less than 750,000 cells per ml. Once changes are made, samples are taken on an accelerated basis and tested. Enough sample results must be less than 750,000 in order to regain “legal” status.

Arithmetic vs. Geometric Means

There often is confusion about somatic cell numbers and scores, how the numbers are computed, and their relevance to milk quality and food safety. References are made to arithmetic means and geometric means. What is the difference between these two types of means (6)?

The two different methods may be used to calculate an “average somatic cell count” when multiple samples have been taken. In the US, for example, if the past three monthly cell counts were 600,000, 400,000 and 500,000, the average would be calculated by arriving at a total and dividing by 3, (1,500,000/3=500,000). This produces an arithmetic mean or average.

A different method, used in Europe and other locations, is used to calculate an average somatic cell count. It is termed the geometric mean. The geometric mean calculation always produces a value somewhat less than the arithmetic mean for the same data set. For the data set used before (600,000, 400,000, 500,000) the geometric mean would be calculated as approximately 493,000. A single high count in a data set has a greater impact on the arithmetic mean than the geometric mean and one very high value isn’t as likely to trigger regulatory action using the geometric mean procedure.

Arithmetic and Geometric Means for Similar Data Sets

Raw cell Counts Raw cell Counts
400,000 300,000
500,000 400,000
600,000 700,000
Arithmetic mean=500,000 Arithmetic mean=467,000
Geometric mean=493,000 Geometric mean=438,000

Herd somatic cell numbers can be the result of testing individual cows and averaging the values or may be based on a bulk tank sample (BTSCC) that represents the whole herd. Individual cow cell counts are typical of DHIA testing programs and are a good measure of the infection status of all cows in the herd. Bulk tank values are indicative of the status of the herd when all the milk produced is blended together in the bulk tank and then sampled. These numbers differ, however, for several reasons.

High producing cows with high cell counts have a greater impact on the BTSCC than a low producing cow with a high SCC simply due to the total volume of milk produced. This is especially true for small herds when a high cell count cow can significantly influence the BTSCC. In a large herd this is less likely to happen because of the dilution effect.

US vs. European SCC Testing for Regulatory Purposes

The US milk quality monitoring system requires that approximately monthly samples, taken from farm bulk milk, be tested for bacteria and somatic cells. When a single bulk tank somatic cell count (BTSCC) exceeds 750,000/ml, it raises a concern. When two of the last four consecutive milk samples are above the limit, the producer is placed on notice and if three of the last 5 are above 750,000/ml the Grade A license is suspended until corrections are made and acceptable values (less than 750,000/ml) obtained. The US does not average several results from a particular time period; rather it uses the individual monthly cell count results.

The European approach differs in terms of how monthly sample results are used. Monthly milk samples are tested for bacteria and somatic cells much like the US program. Europeans then calculate an average of the three monthly values for somatic cells using the geometric mean calculation method. The resulting number becomes the single reference value for the three-month period. If the value is too high (above 400,000/ml in European Union countries), producers are placed on notice but they actually have the subsequent three-month period to get cell counts back in line. After the next three-month period results are averaged, an acceptable value will cancel the notice but a high value will result in license suspension.

A commonly expressed US concern is that the European method of calculation produces a lower number than the US method, so US producers are penalizing themselves. When comparing the two methods for assessing average SCC and ramifications of changing the system, several issues have to be considered (1).

One is the upper limit allowed (750,000 US vs. 400,000 EU). Another is the calculation method (arithmetic vs. geometric) and the third difference is use of an average for a three-month period in Europe as the single reference value for the period. Basically, Europeans have a lower maximum number to attain but more time to get there, based on how values are accumulated and summarized.

Summary of US vs. European SCC Programs

SCC Considerations US EU
Maximum level of SCC allowed in bulk milk 750,000/ml 400,000/ml
SCC value used Consecutive monthly SCC values Geometric mean of three monthly samples used as a single value for the period
Conditions causing producer license to be suspended 3 of 5 consecutive samples test greater than 750,000/ml Second consecutive three month geometric mean

Differences in SCC limits has been a concern for the US dairy industry when attempting to export milk products to Europe. Europe requires the SCC of the raw milk supply used in such products be at 400,000/ml or less based on the geometric mean calculation method. To determine this, the US has to evaluate the raw milk supply used in making the export products, using the geometric mean calculation procedure, and then certify that it had a SCC value less than 400,000/ml. This means extra effort for export products.

It has been well documented that the SCC of individual cows is directly related to ability to achieve maximum production. The message is simple. Keep the SCC as low as possible and it will allow more milk per cow, eliminate the risk of PMO violations and, and provide a better product to milk processors whether used as fluid milk or converted to cheese or other processed products.


  1. Adkinson, R.W., R.H. Gough, R. Graham, and A. Yilmaz. 2001. Implications of proposed Changes in Bulk Tank Somatic Cell Count Regulations. J. Dairy Sci. 84: 370-374.
  2. Ma Y., C. Ryan, D.M. Barbano, D.M. Galton, M.A. Rudan, and K.J.Boor. 2000. Effects of Somatic Cell Count on Quality and Shelf Life of Pasteurized Fluid Milk. J. Dairy Sci. 83: 264-274.
  3. Raubertas, R. F., and G. E. Shook. 1982. Relationship between Lactation Measures of Somatic Cell Concentration and Milk Yield. J. Dairy Sci. 65: 419-425.
  4. Schallibaum, M. 2001. Impact of SCC on the Quality of Fluid Milk and Cheese. National Mastitis Council, Inc. 40th Annual Meeting Proceedings. 38-46.
  5. Schukken, Y.H., G. Bennet, L. Green, T. van Werven. 2001. Can Somatic Cell Counts Get Too Low? National Mastitis Council, Inc. 40th Annual Meeting Proceedings. 19-28
  6. Shook, G. and P. Ruegg. 1999. Geometric Mean Somatic cell Counts: What They Are; What They Do. National Mastitis Council, Inc. 38th Annual Meeting Proceedings. 93-100.
  7. U.S. Department of Health and Human Services, Public Health Service, Food and Drug Administration. Grade -A- Pasteurized Milk Ordinance. 1997 Revision.


Winston Ingalls

Winston Ingalls
17 articles


Read more »



Every day millions of dairy animals are milked, fed and maintained by DeLaval solutions in more than 100 countries worldwide – and DeLaval meets with over 10 000 milk producers on their farms. 

Read more »