- PDF Version (1.5M) - Print Friendly - Help

Proposed regulatory framework for unpasteurised milk products

5 Risk management

5.1 An overview of the risks associated with unpasteurised milk products

5.1.1 Milk as a source of pathogens

Milk that is raw (that is, has not been subjected to pasteurisation or an alternative treatment method) may contain pathogenic bacteria. Pathogens can contaminate the milk either directly, for example if an animal has an infection such as mastitis (clinical or sub-clinical), or through unhygienic practices employed during milk harvesting, milk collection and transport.

Historically, a wide range of pathogens have been associated with milk. For example, the systematic review of evidence undertaken by Massey University and summarised in section 4.6.1 above identified moderate evidence of a casual link between consumption of raw milk (and raw milk products) and Campylobacter spp., E.coli spp., L. monocytogenes, and Salmonella serovars. The level at which such pathogens will be present in raw milk may vary from low to high. On occasions, such levels may lead to overt disease in vulnerable members of the population (such as the very young, pregnant women, the frail elderly and those with weakened immune systems), even though they would not cause clinical illness in healthy individuals.

5.1.2 Reducing the levels of pathogens in milk

As scientific knowledge has advanced and it has been understood that liquid milk can transmit illness, strategies to reduce the incidence of pathogens in milk have been introduced. Such strategies have included improved animal health practices, such as bovine tuberculosis control programmes, somatic cell count monitoring, and more effective sanitation of milking machines. However, even when such strategies are employed, they do not guarantee the elimination of all pathogens from milk. As a result, it is now standard health practice to ensure that liquid milk destined for drinking, especially by vulnerable consumers, is heat treated, for example by pasteurisation, ultra heat treatment or scalding³. Pasteurisation removes most pathogens of concern and reduces the number of spoilage organisms present. Pasteurised milk will still spoil even under refrigeration, due to the growth of any spoilage organisms remaining after heat treatment, unless it is subjected to a process such as ultra heat treatment and aseptic packaging.

5.1.3 Extending the shelf life of milk by converting it into dairy products

Traditionally, milk has been converted into cheeses and other fermented products to extend its shelf life. For most of these milk products, processing commences with the addition of specific cultures which cause the milk to ferment. Dependent on the type of product being made, the shelf life of milk can in this way be extended from weeks (as occurs with yoghurt), to months or even years (as in the case of hard cheeses). The length by which the shelf life is extended is dependent on the intrinsic characteristics of the product, the processing techniques employed, and the inhibiting effect which these have on the growth of micro-organisms that cause spoilage. For example, characteristics such as acidity, low moisture content, and antimicrobial factors produced by the fermenting organisms all result in the elimination or growth inhibition of both micro-organisms that cause spoilage and of pathogens that were in the milk. The final spoilage patterns that result are due to the final product characteristics brought about by all of these factors. For example, the shelf life of a dry, hard cheese such as parmesan is longer than that of a very moist product such as yoghurt.

5.1.4 Raw milk in cheese making

Cheeses that are made from poor quality milk may display flavour and quality defects due to the presence of contaminating micro-organisms. As bacteria are present in raw milk and can grow rapidly, the quality of raw milk can soon deteriorate unless it is processed soon after collection. Cheeses made from raw milk can show varying characteristics from day to day, due to the different micro-organisms that can be present. Many cheese makers prefer to pasteurise milk before they produce cheese as this can result in a more consistent product and reduce the potential for production batches to fail requirements for flavour, quality and food safety. However, other cheese makers value the flavour, texture and quality characteristics which they consider the use of raw milk can produce in a cheese.

During the early stages (warm and moist) of cheese making, any pathogenic and spoilage bacteria present may multiply unless a preliminary heating step has destroyed them. However as the starter culture grows and starts to increase acidity in conjunction with the release of inhibitory factors, then the growth of the pathogens present will be slowed. In the next stage of cheese making – curd making – the growth of pathogens will in most cases have ceased. Whilst at the beginning of this stage a greater number of pathogens may now be present than were in the incoming milk, the process of producing curd, reducing the moisture content, salting, and ripening will usually result in pathogen numbers decreasing to low levels. However, this expected die-off of pathogens takes time and may not occur if the cheese making process does not include a lengthy ripening stage of at least several weeks duration (an example of products not subject to lengthy ripening are the Mexican-style unripened soft cheeses), or if the factors that are inhibitory to pathogens are reversed, such as occurs when the acidity of a product decreases.

The inhibitory effects during cheese making have been studied and some examples of this scientific evidence are described below:

• Certain soft cheeses, such as Camembert (where the pH is almost back to neutral at the end of the ripening phase), and Quesco Blanco and Quesco Fresca (which have a high moisture content and a neutral pH) will support the growth of L. monocytogenes. Quesco Blanco and Quesco Fresca have been associated with outbreaks of listeriosis (CDC 2001⁴; Linnan et al 1989)⁵.

• Hard cheeses such as Cheddar and Parmesan have acid pHs and low water activity and do not support the growth of listeria. They show a marked decline in live bacteria during ripening (Yousef AE and Marth, EH 1990⁶).

• While pathogen population levels may not decline to undetectable levels in hard cheeses during aging, pathogen survival is strongly influenced by the moisture content and pH levels, and there have been no recorded outbreaks of illness associated with cheeses aged for a minimum of 60 days (ILSI, 2005⁷).

• In the United States of America, 11 events between 2000 and 2005 (five relating to L. monocytogenes, two to Salmonella and two to M. bovis), have been linked to the consumption of cheeses made from unpasteurised or raw milk. However, ten of these events were linked to Mexican-style soft cheeses, which in most cases were purchased from street vendors rather than retail stores, meaning that it is not clear whether the products were made following Good Hygiene Practice (Quoted by the US FDA).

5.1.5 Risk assessments of unpasteurised milk products

A full quantitative risk assessment of milk products requires the availability of extensive data relevant to each stage of manufacture to allow modelling of the impact of changes in parameters such as pH, the likelihood of the survival of contaminating micro-organisms, and the subsequent potential of products to cause illness. However, in general, there is a scarcity of such data and, for this reason, very few full quantitative risk assessments have been undertaken for individual cheeses either in New Zealand or overseas. One assessment that has been done was undertaken by NZFSA to support the risk management decision taken with regards to the importation of the French unpasteurised sheep milk cheese, Roquefort, refer to: http://www.nzfsa.govt.nz/imported-food/imports-portfolio/risk-management-roquefort/riskmanagementdecision_roquefort.pdf

As there are many types of cheeses, probably several thousand, each of which has its unique processing parameters, it is not possible to complete a general risk assessment that includes all unpasteurised milk cheeses or other types of unpasteurised milk products. However, it is possible to group dairy products according to some key characteristics, although the boundaries between the groups may not always be clear-cut. For example, several national and international agencies have conducted risk assessments on the association of groups of dairy products with L. monocytogenes, and the outcome of one such assessment by the US FDA is described in Section 5.1.6 below.

5.1.6 What we do know about the risks associated with milk products

Despite the limitations outlined above, it is possible to identify three general levels of risks for milk products (and other dairy products), using evidence collected from challenge studies and outbreaks of illness associated with the consumption of dairy products.

Assuming that milk may on occasion contain pathogens, then the potential risk to general members of the public of illness from consumption of milk and milk products will be:

• extremely low if the milk is pasteurised or subjected to an equivalent treatment;

• low if the milk is treated in a way that results in a combination of extrinsic and intrinsic factors that minimise the survival and growth of the pathogens; or

• moderate to high if there are no factors or limited factors (either extrinsic or intrinsic) that inhibit the survival and growth of the pathogens.

The regulatory framework for unpasteurised milk products developed by NZFSA, including the three descriptions of unpasteurised milk products outlined in Section 6.1, is based on these three general levels of risk.

Risk assessments which support the proposed NZFSA regulatory framework include those undertaken by the US FDA in 2003 relating to listeria in ready to eat foods. These ranked unpasteurised fluid milk as high risk, soft unripened cheeses as moderate risk, and other cheeses as low risk, with hard cheese being ranked the lowest risk to the general population of all the products in the 23 categories of ready to eat foods studied.

Outbreaks of illness have rarely been associated with unpasteurised milk cheeses, but nevertheless the potential for such events exist (as the Massey University systematic review of evidence demonstrates). It is therefore necessary to endeavour to ensure that there is minimal pathogen contamination of raw milk used in such products, and that the subsequent processing of the milk ensures that the growth and survival of any pathogens that may be present is minimised by stringent control of the manufacturing process. This means that the manufacture of unpasteurised milk products needs to be subject to rigorous controls and that all those involved understand the specific hazards and risks, and how to control them.

In the draft specifications in Appendix 1, controls and requirements proposed by NZFSA for raw milk products are outlined. In developing these, NZFSA has taken account of the work undertaken by Codex and published in the Codex Code of Hygienic Practice for Milk and Milk Products CAC/RCP 57-2004. This Code contains specific additional provisions for raw milk products. Particular emphasis is given to on-farm practices to ensure the quality of the raw milk, and to strict controls during processing to ensure factors essential for the manufacture of a safe product are monitored and managed effectively. The Codex Code can be accessed at:

http://www.codexalimentarius.net/download/standards/10087/CXP_057e.pdf

5.2 Managing the hazards associated with unpasteurised milk products

The framework that NZFSA proposes utilises Good Hygienic Practice (GHP) and hazard-based controls to ensure that, at the end of manufacturing, unpasteurised milk products comply with specific food safety criteria. It is proposed that there will be specifications which contain requirements for the production of some unpasteurised milk products in New Zealand. Standards under the Food Act will allow for the importation of an equivalent range of products and the import requirements that apply. NZFSA will develop guidance to aid producers, evaluators, verifiers and importers to determine which requirements apply.

NZFSA will also develop a risk communication strategy to manage public understanding of the higher likelihood of the presence of hazards in unpasteurised milk products. Unlike some overseas countries, New Zealand does not have a culinary tradition of consuming unpasteurised milk products. As a result the New Zealand public’s awareness of the hazards associated with the consumption of such products is limited. This is borne out by the survey of consumers commissioned by NZFSA, which found that almost a third of the general public surveyed considered that raw or unpasteurised milk cheeses/products were, as a general group, as safe as pasteurised milk products. The NZFSA future risk communication strategy will be targeted at those consumers most likely to be vulnerable to any hazards associated with the unpasteurised milk products that would be available in New Zealand – namely the young, frail elderly, pregnant and immune-compromised (sometimes referred to as ‘YOPIs’).

To support the risk communication strategy and assist consumer recognition of unpasteurised milk products, such products would also need to be labelled as containing raw or unpasteurised milk.

5.2.1 Risk communication

Risk communication is an important component of managing risk. Risk communication involves providing people with information about a risk or hazard (including putting that risk or hazard into context) and allowing them to make an informed decision about their actions in relation to that risk or hazard. It can be described as ‘alerting’ and ‘reassuring’.

NZFSA runs a substantial programme of informing consumers about risks and hazards from food. This includes a range of information brochures, fact sheets, a comprehensive web site, and working with public health and industry organisations via the Foodsafe partnership. Brochures are distributed to the public through health professionals, public health units, territorial authorities, and directly from NZFSA.

Some of NZFSA’s information brochures target specific audiences and some are issue-based. Examples of brochures for audiences with specific food safety needs include Food safety when you have a low immunity aimed at the very young, frail elderly, pregnant and immune-compromised groups; Food safety in pregnancy; and Good food/Safe food for older people. Issue-based publications include Meet the Bug (which describes foodborne pathogens) and Agricultural compounds in food. A full list of NZFSA consumer information brochures is available on the NZFSA web site at: http://www.nzfsa.govt.nz/consumers/index.htm.

When NZFSA updates its advice on a topic, or when a new food safety risk or hazard is communicated to New Zealand consumers, relevant new information is added immediately to the web site and to existing publications when they are next reviewed. Depending on the specific risk or hazard, a new publication might be developed.

5.2.2 Risk communication for unpasteurised milk products

In 2007 New Zealand’s food laws were changed to allow the direct importation of some unpasteurised milk cheeses not previously available (i.e. Roquefort and some Italian hard grating cheeses). Part of the risk management plan developed at the time by NZFSA included the design of a point-of-sale flyer for Roquefort and a more general flyer about raw milk products for distribution to medical offices. The term ‘raw milk’ has been used here for information produced to date as this is the terminology that was used in these publications.

Two new publications for at-risk groups were also released at that time which included new information alerting them to the need to avoid eating products made from raw milk.

Since 2007, NZFSA’s brochure for pregnant women has been updated to include the raw milk message. The NZFSA brochure for YOPI groups is also currently being updated.

If this proposal proceeds, and a wider range of unpasteurised milk products become available on the New Zealand market, then NZFSA will further develop its risk communication strategy for such products. Details of the type of adjustments that would be made to the strategy are given in Section 9.3.

5.2.3 Labelling

In the first discussion paper, NZFSA identified labelling of unpasteurised milk products as a possible tool that could be used to mitigate any hazards associated with such products.

The labelling requirements for all food sold in New Zealand (including dairy products) are set out in the Code. Among other things, the Code requires that the label on a package of food for retail sale includes the name of the food and a list of its ingredients. Standard 1.2.4 of the Code requires that ingredients must be declared using the common name of the ingredient, or a name that describes the true nature of the ingredient, or if applicable a generic name. This requirement means that in relation to products made from raw or unpasteurised milk, the ingredient declaration would need to include a statement that the milk is ‘raw’ or ‘unpasteurised’. In the case of products made other than from cows’ milk, labels would also need to include the common name of the species from which the milk is sourced.

Feedback received on the first discussion paper indicated that submitters supported labelling of unpasteurised milk products as an effective means of helping consumers to recognise such products. The NZFSA survey of consumer awareness of unpasteurised milk products also identified labelling as a valuable mechanism for ensuring consumers are made aware when they are buying products made from raw or unpasteurised milk. Some consumers surveyed also considered that an explanation and/or warning of any hazards associated with unpasteurised milk products should be included on labelling.

The Code does allow for mandatory advisory statements to be used when consumption of a food exposes the general population, or a population subgroup, to a health and safety risk, or where guidance about a food is needed to maintain public health and safety. Such mandatory advisory statements are included separately, and in addition to, ingredient lists. The list of foods that are currently required by the Code to have a mandatory advisory statement includes unpasteurised milk and liquid milk products, but not unpasteurised milk cheeses or other unpasteurised milk products. This means that the labelling on unpasteurised milk and liquid milk products (which are available in some Australian states) must include an advisory statement ‘to the effect that the product has not been pasteurised’, but that no such mandatory advisory statements nor any other warnings are currently required on the labels of unpasteurised milk cheeses or other unpasteurised milk products.

Like New Zealand, Australia is considering liberalising its regulatory environment for unpasteurised milk products. While the likely outcome is an Australian only processing standard, labelling matters will be addressed and, under the terms of the Agreement Between the Government of Australia and the Government of New Zealand Concerning a Joint Food Standards System (the Food Treaty), any changes to labelling are to apply jointly in Australia and New Zealand. NZFSA will therefore continue to collaborate and consult with FSANZ to ensure that the needs of New Zealanders are addressed. NZFSA’s preferred option is that all (i.e. not just liquid) raw milk products that are subject to the proposed new specifications (see Section 6.1. below) should be labelled with a mandatory advisory statement to the effect that they contain ‘unpasteurised milk’.

If agreed, any such mandatory advisory statement would be introduced jointly in New Zealand and Australia at an appropriate future time that would equate with FSANZ’s timeline for completing its work on unpasteurised milk products (gazettal of any new Australian requirements is currently set for February 2011⁸). NZFSA favours the use of the term ‘unpasteurised milk’ rather than ‘raw milk’ in any future mandatory advisory statement because this reflects submissions on the first NZFSA discussion document, and the market research commissioned by NZFSA to determine consumer understanding of commonly used terms relating to raw or unpasteurised milk products.

If adopted in future, such a mandatory advisory statement would also apply to raw milk products that are not required to bear a label, for example those made and packaged on the premises from which they are sold, as the statement would need to be displayed on, or in conjunction with, the display for the food concerned, or provided to the purchaser on request.

In the meantime, producers of unpasteurised milk products may wish to voluntarily label their products with a mandatory advisory statement. Recommendations relating to labelling will be incorporated in the Code of Practice relating to raw milk products that NZFSA proposes to develop for industry operators (refer Section 6.2.2.).

³ It should be noted that this proposal does not put forward any alternative to this standard health practice for raw milk intended for drinking and, apart from the very limited farm gate sales of raw milk already permitted under the Food Act, there is no intention to allow raw drinking milk to be sold in New Zealand.

⁴ CDC (2001) Outbreaks of listeriosis associated with homemade Mexican-style cheese. Morbidity and Mortality Weekly Report 50, 560-562.

⁵ Linnan, MJL et al (1989) Epidemic listeriosis associated with homemade Mexican-style cheese, New England Journal of Medicine, 319, 823-828.

⁶ Yousef AE and Marth EH (1990) Fate of Listeria monocytogenes during the manufacture and ripening of Parmesan cheese. Journal of Dairy Science 73, 3351-3356.

⁷ ILSI (2005), Achieving continuous improvement in reductions in foodborne listeriosis – a risk-based approach. Journal of Food Protection 68, 1932-1994.

⁸ For further information about the FSANZ work plan, visit: http://www.foodstandards.govt.nz/_srcfiles/Work%20Plan%20LATEST4.pdf