Food Standards Scotland (FSS) has published a report that it says can be used as a basis to improve safety for artisan Scottish producers making cheese from unpasteurized milk.

The literature review, by Catherine Donnelly, provides evidence to Scottish artisan cheesemakers and enforcement officials in managing the microbiological safety of artisan cheeses and those produced from unpasteurized milk.

Donnelly, an expert on the microbiological safety of food, believes that making cheesemakers use pasteurized milk is not the best way to produce safe cheeses, She contends it is better to educate them on how to improve the safety of raw milk products.

Contamination of raw milk, and products made with it, by pathogens cannot be eliminated despite efforts to control milk hygiene. Consequently, cheesemakers must implement a range of strict controls to ensure a safe product.

Microbiological safety of cheese made from unpasteurized milk is dictated by the safety of raw milk used for its production. Improving milk hygiene can be done through herd management, mastitis control, a focus on feeding regimes, and sanitation during milking, storage and transportation to cheesemakers.

The report outlines the evidence on food safety controls to reduce the risks of food poisoning bacteria in the production of raw milk cheeses. These controls include critical control points (CCPs), predictive modeling, and validation and verification of firms’ processes such as the use of microbiological testing data.

To achieve process control that assures consistent production of microbiologically safe and high quality cheeses, cheesemakers must routinely monitor microbiological and physiochemical results from milk and cheese testing. Microbiological and physicochemical data together should help verify a HACCP plan is operating as intended.

Relevant pathogens
The main pathogens relevant to cheeses made from unpasteurized, raw milk are Listeria monocytogenes, Shiga toxin-producing E. coli (STEC), Salmonella and Staphylococcus aureus. The United Kingdom regards any STEC as potentially pathogenic and does not recognize specific serotypes as more or less pathogenic. It considers the presence of STEC in food confirmed when stx genes are detected in an isolated E. coli strain.

One of the highest profile cases that FSS has been involved in featured Dunsyre Blue cheese made by Errington Cheese from unpasteurized cow’s milk. The cheese was linked to an E. coli O157 outbreak in 2016 that sickened 26 people. Seventeen of the patients required hospitalization and a 3-year-old girl died. The agency concluded the cheese was the source of the outbreak, but the company has always denied the link. This year, Errington Cheese was cleared of breaching food hygiene regulations.

Microbiological data provided by the Specialist Cheesemakers Association (SCA) showed lower prevalence of coagulase-positive staphylococci and generic E. coli in UK raw milk for cheese making compared to surveys in the United States. No Salmonella or E. coli O157:H7 were found in 298 and 225 samples analyzed. Listeria monocytogenes was detected in 43 of 639 samples, although this included results of re-sampling after detection.

In the United Kingdom, there have been eight outbreaks linked to unpasteurized milk cheese since 1983 and 53 outbreaks globally (Yoon et al. 2016; Fox et al. 2017). Of the eight U.K. outbreaks, six were attributed to E. coli O157 and one each to Salmonella and Staphylococcus spp. Although not implicated in an outbreak connected to cheeses from unpasteurized milk, Listeria monocytogenes can be isolated from such cheese.

The report recommends that during epidemiological investigations of outbreaks involving cheese, collection of compositional data such as pH levels, salt and moisture content, could reveal information about causative factors, including lack of process control.

Assuring safety of cheese made from raw milk is influenced by use of raw milk of high microbiological quality with low pathogen contamination levels; rate and degree of acidification achieved during cheese making; rate of pathogen inactivation during cheese making and ageing and prevention of recontamination from the processing environment or at retail.

For each cheese type produced in Scotland, consideration must be given to a number of factors, including raw milk testing, including milk filters; raw milk handling, storage and transportation; environmental monitoring and control of the cheese making environment; cleaning and sanitation; and good manufacturing practices (GMPs).

Possible predictions
A number of predictive models are available to help cheesemakers make safety assessments by predicting their growth potential in cheese but most over-predict pathogen growth in raw milk cheese.

The Raw Milk Cheese (RMC) Decision Support Tool is the most appropriate model for cheesemakers to determine the microbiological safety of their process and finished products. It was developed by the Australian Specialist Cheesemakers’ Association and Dairy Food Safety Victoria (DFSV).

Since most raw milk cheeses are produced by small manufacturers, there is likely to be variability in processing from batch to batch and such variability must be considered in challenge studies so the range of process efficacy can be determined. Surrogate organisms must possess the same characteristics as the pathogen, such as acid, salt or heat tolerance. Reasons including cost mean it is impractical for most artisan cheese producers to do challenge studies.

Challenge studies have shown the microbiological risk of cheeses made from unpasteurized milk can be reduced by a proper acidification, ripening maturation process, and constant monitoring of the hygiene environments for milk production, cheese making and the post-manufacturing stage.

Impact of cheese type
As of August 2018, of 64 cheese types being produced in Scotland, 20 were manufactured from unpasteurized milk from cows, sheep and goats. Twenty-five of the 64 cheeses are hard cheese varieties and 22 are soft. only one of the 22 soft varieties is made from unpasteurized milk, versus 12 of the 25 hard varieties made from unpasteurized milk.

Cheese classification based on pH and Aw is useful for the assessment of risk of survival of STEC, Listeria monocytogenes and other pathogens but alternatives that account for processing steps wher pathogens can be introduced or reduced should also be considered.

Whilst sale of raw drinking milk is permitted in England and Wales, sale of it and unpasteurized cream has been banned in Scotland since 1983 following milk-related illnesses and 12 potentially associated deaths. There are no restrictions on the sale of raw milk cheeses in Scotland, provided they have been produced in compliance with European unio food hygiene regulations.

STEC contamination of cheese can be prevented through milk hygiene and prevention of fecal contamination of milk and cheese. While surveys document a low prevalence of STEC in tested cheese, challenge testing studies show potential for survival during manufacture of a range of cheese types.

The report also covers bacteriophages that have been used to control foodborne pathogens such as Listeria monocytogenes in cheeses. Cheesemakers considering their use should be aware of the potential limitations such as phage resistance associated with this strategy.

Advising artisan producers making raw milk cheeses to eliminate silage feeding in favor of dry hay or pasture feeding shows promise to reduce potential presence of Listeria monocytogenes, and potentially other pathogens, in milk used for cheese making.

The report recommends providing one-on-one technical assistance to small scale cheesemakers to aid in development of individual Risk Reduction Protocols including guidance on validation and a survey of the microbiological quality of raw milk intended for artisan cheese production in Scotland.