Food safety scheme and risk assessment

Food Safety Risk
Assessment of NSW Food
Safety Schemes

March 2009

Contents
Executive summary …………………………………………………………………………………………….. 6 Introduction…………………………………………………………………………………………………….. 12 Dairy food safety scheme …………………………………………………………………………………… 16 Meat food safety scheme ……………………………………………………………………………………. 39 Plant products food safety scheme ……………………………………………………………………….. 67 Seafood safety scheme ………………………………………………………………………………………. 85 Vulnerable persons food safety scheme ……………………………………………………………….. 105 Egg food safety scheme …………………………………………………………………………………… 120 Risk assessment – Conclusion ……………………………………………………………………………. 137 Appendix 1: Microbiological and chemical hazards of concern …………………………………… 138 Appendix 2: Australian food recalls (2004–2008) ………………………………………………….. 161 Appendix 3:
Australian foodborne illness outbreaks (1995–2008) ………………………………. 164

Food Safety Scheme Risk Assessment

Page 1 of 189

Tables
Dairy food safety scheme
Table 1 – Pathogenic microorganisms detected in raw milk………………………………………… 17 Table 2 – Microbiological hazards in dairy products ………………………………………………….. 18 Table 3 – Consumption of dairy products ……………………………………………………………….. 21 Table 4 – Summary of foodborne illness outbreaks attributed to dairy products and foods including dairy as an ingredient……………………………………………………………………………. 23 Table 5 – NZFSA risk profile outcomes examining hazards in dairy products ………………….. 28 Table 6 – Risk ranking for dairy products contaminated with Listeria monocytogenes ………. 29 Table 7 – Risk ranking of dairy products ………………………………………………………………… 30 Table 8 – Microbiological hazards in livestock and poultry ………………………………………….. 41 Table 9 – Consumption of meat and meat products in Australia ………………………………….. 45 Table 10 – Consumption of processed meats in Australia …………………………………………… 45 Table 11 – Summary of foodborne illness outbreaks attributed to all meat, meat products and meat included as an ingredient (1995–2008) (including poultry, game meat and processed meat products) ………………………………………………………………………………………………… 46 Table 12 – Prevalence of microbiological hazards in Australian beef and sheep meat……….. 48
Table 13 – Prevalence of microbiological hazards on retail chicken meat in NSW (2005–06) 50 Table 14 – Foodborne illness outbreaks of listeriosis from processed meats …………………… 52 Table 15 – Prevalence of Listeria monocytogenes in processed meats ………………………….. 52 Table 16 – NZFSA risk profile outcomes examining hazards in meat …………………………….. 54 Table 17 – Risk ranking for meat and meat products ………………………………………………… 55 Table 18 – Risk ranking for processed poultry meat products ……………………………………… 57 Table 19 – NZFSA risk profile outcomes examining hazards in poultry meat …………………… 57 Table 20 – NZFSA risk profile outcomes examining hazards in processed meats ……………… 60 Table 21 – Risk ranking for processed meat products ……………………………………………….. 61 Table 22 – Risk ranking for L. monocytogenes-contaminated processed meats ………………. 62 Table 23 – Microbiological hazards associated with plant products ………………………………. 67 Table 24 – Consumption of fruits and vegetables in Australia ……………………………………… 71 Table 25 – Summary of foodborne illness outbreaks attributed to plant products ……………. 74 Table 26 – Risk ranking for plant products contaminated with Listeria monocytogenes …….. 78 Table 27 – Summary of international hazard identification studies for seafood ……………….. 85 Table 28 – Hazards in seafood and seafood products ……………………………………………….. 86 Table 29 – Production volumes for seafood in Australia and NSW 2006–07 ……………………. 88 Table 30 – Consumption of fish and seafood products in Australia ………………………………. 90 Table 31 – Failure rate for imported seafood products (1998 – 2003) …………………………… 90 Table 32 – Summary of Australian seafood testing results …………………………………………. 91 Table 33 – Summary of high mercury levels in NSW seafood ……………………………………… 91 Table 34 – Prevalence of L. monocytogenes in UK retail smoked fish …………………………… 92 Table 35 – Summary of foodborne illness outbreaks attributed to seafood …………………….. 94 Table 36 – Risk ranking for seafood products
contaminated with Listeria monocytogenes …. 99 Table 37 – Seafood consumption required to reach reference doses for methylmercury …. 101 Table 38 – Summary of foodborne illness outbreaks attributed to food served to vulnerable persons ………………………………………………………………………………………………………… 109

Food Safety Scheme Risk Assessment

Page 2 of 189

Table 39 – Institutional foodborne illness outbreaks as a percentage of all outbreaks …….. 110 Table 40 – Relative susceptibility to listeriosis for different sub-groups ……………………….. 113 Table 41 – Estimated cases of listeriosis for vulnerable population sub-groups for each food category, based on US data ………………………………………………………………………………. 114 Table 42 – Hazards in the production of shell eggs and egg products …………………………. 121 Table 43 – Prevalence of chemical residues in eggs ……………………………………………….. 123 Table 44 – Prevalence of Salmonella in Australian eggs …………………………………………… 125 Table 45 – Consumption of eggs and egg products in Australia …………………………………. 125 Table 46 – Summary of foodborne illness outbreaks attributed to eggs, egg products and eggs used as an ingredient ……………………………………………………………………………….. 126 Table 47 – Risk ranking for type and use of eggs …………………………………………………… 129 Table 48 – Top Salmonella serovars from major sources …………………………………………. 140 Table 49 – Characteristics of Salmonella ………………………………………………………………. 140 Table 50 – Characteristics of Campylobacter
…………………………………………………………. 141 Table 51 – Characteristics of Staphylococcus aureus ………………………………………………. 142 Table 52 – Characteristics of Clostridium perfringens ………………………………………………. 143 Table 53 – Characteristics of Bacillus cereus …………………………………………………………. 144 Table 54 – Characteristics of Listeria monocytogenes ……………………………………………… 146 Table 55 – Characteristics of Vibrio parahaemolyticus……………………………………………… 148 Table 56 – Characteristics of Shigella spp. ……………………………………………………………. 149 Table 57 – Characteristics of pathogenic Escherichia coli …………………………………………. 151 Table 58 – Characteristics of Clostridium botulinum………………………………………………… 152 Table 59 – Characteristics of Yersinia enterocolitica………………………………………………… 153 Table 60 – Important Aspergillus, Fusarium and Penicillium species and their mycotoxins . 158 Table 61 – Recalls of dairy products between 2004 and 2008 …………………………………… 161 Table 62 – Recalls of meat products between 2004 and 2008 …………………………………… 162 Table 63 – Recalls of plant products between 2004 and 2008 …………………………………… 163 Table 64 – Recalls of seafood products between 2004 and 2008 ……………………………….. 163 Table 65 – Foodborne illness outbreaks attributed to milk, dairy products and dairy products used as an ingredient ………………………………………………………………………………………. 165 Table 66 – Foodborne illness outbreaks attributed to meat, meat products and meat products used as an ingredient ………………………………………………………………………………………. 166 Table 67 – Foodborne illness outbreaks attributed to plant products ………………………….. 174 Table 68 – Foodborne illness outbreaks attributed fish and seafood products ……………….
175 Table 69 – Foodborne illness outbreaks attributed to foods served to vulnerable persons .. 182 Table 70 – Foodborne illness outbreaks attributed to eggs, egg products and eggs used as an ingredient ……………………………………………………………………………………………………… 185

Food Safety Scheme Risk Assessment

Page 3 of 189

Abbreviations
aw

Water activity

ABARE

Australian Bureau of Agricultural and Resource Economics

ABS

Australian Bureau of Statistics

ACMF

Australian Chicken Meat Federation

ACMSF

Advisory Committee on the Microbiological Safety of Food (UK)

AECL

Australian Egg Corporation Limited

ANZDAC

Australia New Zealand Dairy Authorities Committee (formerly
ADASC)

AMRA

Australian Milk Residue Analysis Survey

APL

Australian Pork Limited

APVMA

Australian Pesticide and Veterinary Medicines Authority

ASP

Amnesic Shellfish Poisoning

ASQAP

Australian Shellfish Quality Assurance Program

ATDS

Australian Total Diet Survey/Study

APVMA

Australian Pesticides and Veterinary Medicines Authority

AQIS

Australian Quarantine and Inspection Service

BSE

Bovine Spongiform Encephalopathy

BTEC

Brucellosis and Tuberculosis Eradication Campaign

CAC

Codex Alimentarius Commission

cfu

Colony forming unit

CFR

Code of Federal Regulation (US)

CJD

Creutzfeldt-Jakob Disease

DAFF

Department of Agriculture Fisheries and Forestry (Australian Government) (formerly AFFA)

DFSV

Dairy Food Safety Victoria

DSP

Diarrhoetic Shellfish Poisoning

EFSA

European Food Safety Agency

EHEC

Enterohaemorrhagic E. coli

ERL

Extraneous Residue Limit

EU

European Union

FAO

Food and Agricultural Organization of the United Nations

FDA

Food and Drug Administration (US)

FRDC

Fisheries Research and Development Corporation

FRSC

Food Regulation Standing Committee

FSA

Food Science Australia

FSAI

Food Safety Authority of Ireland

FSANZ

Food Standards Australia New Zealand (formerly ANZFA)

FSIS

Food Safety and Inspection Service (US)

GAP

Good Agricultural Practices

GBR

Geographical BSE Risk

GHP

Good Hygienic Practices

GMP

Good Manufacturing Practices

Food Safety Scheme Risk Assessment

Page 4 of 189

HACCP

Hazard Analysis Critical Control Point

HAV

Hepatitis A virus

HTST

High Temperature Short Time pasteurisation

HUS

Haemolytic Uremic Syndrome

ICMSF

International Commission on Microbiological Specifications for Foods

JECFA

Joint FAO/WHO Expert Committee on Food Additives

KP

Kanagawa phenomenon

MAP

Modified atmosphere packaging

MeHg

Methylmercury

ML

Maximum Level

MLA

Meat & Livestock Australia

MMWR

Morbidity and Mortality Weekly Report

MRL

Maximum Residue Limit

NARM

National Antibacterial Residue Minimisation program

NEPSS

National Enteric Pathogen Surveillance Scheme

NGSP

National Granuloma Submission Program

NRS

National Residue Survey

NZFSA

New Zealand Food Safety Authority

OC

Organochlorine

OP

Organophosphate

PHLS

Public Health Laboratory Service, UK

PIRSA

Primary Industries and Resources South Australia

PISC

Primary Industries Standing Committee

PSP

Paralytic Shellfish Poisoning

PTWI

Provisional Tolerable Weekly Intake

REPFEDS

Refrigerated processed foods of extended durability

RIRDC

Rural Industries Research and Development Corporation

RIS

Regulatory Impact Statement

RTE

Ready-to-eat

SARDI

South Australian Research and Development Institute

SSOP

Sanitation Standard Operating Procedures

STEC

Shiga toxigenic E. coli

SWG

Sector Working Groups

TFAP

Tuberculosis freedom assurance program

TVC

Total Viable Count

UCFM

Uncooked comminuted fermented meats

UHT

Ultra Heat Treated

USDA

US Department of Agriculture

WHO

World Health Organization

YMT

Yolk Mean Time

Food Safety Scheme Risk Assessment

Page 5 of 189

Executive summary
The NSW Food Regulation 2004 contains food safety schemes that outline the regulatory requirements for dairy, meat, plant products, seafood businesses and businesses serving food to vulnerable persons in NSW. A draft egg food safety scheme is currently being finalised for inclusion in the Regulation. The regulatory requirements in the food safety schemes have been introduced over a number of years, either by the NSW Food Authority or its predecessor organisation SafeFood Production NSW. Dairy and meat food safety schemes were carried over from previous legislation. Individual risk assessments
were carried out prior to the introduction of both the seafood and plant products food safety schemes. The development of the vulnerable persons food safety scheme occurred following the introduction of Standard 3.3.1 – Food Safety Programs for Food Service to Vulnerable Populations of the Australia New Zealand Food Standards Code (Food Standards Code). In respect to the egg food safety scheme, the NSW Food Authority conducted a risk assessment prior to developing requirements for the scheme. Within each sector covered by the food safety schemes there are a wide variety of hazards that may potentially be present and cause illness in the consumer. The degree of illnesses caused by these hazards can range from mild illness through to severe and life threatening disease. In general, it is the microbiological hazards associated with foods that are considered more significant, as chemical and physical hazards are rarely detected in food.

This risk assessment document summarises the known information from previous risk assessments, risk profiles and hazard assessments, and includes new or updated information where it is available and applicable to food businesses in NSW.

Dairy food safety scheme
In 2006–2007 there were 684 million litres of milk sold in NSW and ACT, with the average person consuming in excess of 100 L of pasteurised milk each year. A wide variety of bacteria may be present in raw milk with the microbial status of milk being influenced by animal health, the farm environment and production methods. Pasteurisation was successfully introduced to eliminate tuberculosis and brucellosis from milk and nowadays the main microbiological hazards associated with milk and dairy products include Salmonella, Listeria monocytogenes, pathogenic Escherichia coli, Staphylococcus aureus, Campylobacter spp., Yersinia enterocolitica and Cronobacter sakazakii.

Between 1995 and 2008 there were fourteen Australian outbreaks attributed to dairy products, none occurring in NSW. Nine of these outbreaks were associated with consumption of unpasteurised milk. Internationally, foodborne outbreaks associated with dairy products have been attributed to
the use of unpasteurised milk, contaminated non-dairy ingredients, faulty pasteurisation process and poor hygiene. Controlling the safety of milk and dairy products relies on using raw materials (milk and non-dairy ingredients) of good quality, ensuring correct formulation, effective processing, prevention of recontamination and maintenance of temperature throughout the cold chain. Dairy products identified as high risk include unpasteurised milk, soft cheese, dairy desserts, fresh cheeses and dairy dips, as these products may support the growth of pathogenic microorganisms. Two critical steps in controlling pathogens in milk and dairy products are effective pasteurisation, followed by good manufacturing practices to ensure postFood Safety Scheme Risk Assessment

Page 6 of 189

pasteurisation contamination does not occur. Food safety programs targeting these controls have been an effective mechanism for controlling microbial hazards in milk and dairy products.
There are several potential sources of chemical contamination associated with milk production including agricultural and veterinary chemicals, environmental contaminants and chemicals from animal feed. The implementation of on-farm food safety programs has managed these risks, and the risk is considered low as surveys of dairy products have not detected significant levels of chemicals in Australian milk and dairy products.

Meat food safety scheme
It has been estimated that Australians each consume 38.2 kg of beef and veal, 11.4 kg of lamb, 2.7 kg of mutton and 14.4 kg of bacon and ham products each year. Various microbiological hazards are associated with different types of meats, with Salmonella, pathogenic E. coli, Clostridium perfringens, Campylobacter jejuni and the parasite Toxoplasma gondii associated with beef and sheepmeat, while the primary pathogen of concern in pigmeat is Yersinia enterocolitica.

Livestock and poultry can serve as a reservoir for pathogenic microorganisms and within the abattoir environment these pathogens can be transferred from
the gut to the external surfaces of the carcase and contaminate equipment and workers. Currently the risk associated with red meat is considered low, due to the control measures implemented by the meat industry, such as the Australian Standards for the hygienic production of meat and game meat. However, cross contamination of ready-to-eat (RTE) foods by raw meat is considered a potential issue of concern. It has been estimated that if the cross contamination rate of Salmonella increased from 1% to 10% there would be an extra 5000 cases of foodborne illness, while an increase to 50% would result in more than 29,000 cases of salmonellosis across Australia each year.

Poultry is the most widely consumed meat in Australia, with each person consuming 39.5kg of poultry each year. The primary hazards of concern in poultry meat are Salmonella and Campylobacter spp. Contamination of poultry can occur on farm through breeding stock, contaminated water, litter, insects, rodents, wild birds and farm workers. Surveys have identified poultry meat as a significant source of foodborne illness, with 46 confirmed outbreaks and 1170 cases of illness between 1995 and 2002. Due to significant under-reporting, it is estimated that cases of foodborne illness due to processed chicken products may be as high as 79,000 cases per year. A through chain approach is the preferred option to reduce contamination of poultry meat, with estimates that this could reduce levels of poultry-related foodborne illnesses by between 74% and 93% each year.

Processed meat products have also been identified as high risk, with the pathogens of concern including pathogenic E. coli, Salmonella and L. monocytogenes. There have been a large number of recalls of processed meats due to L. monocytogenes, and a significant number of foodborne illness outbreaks. Between 1991 and 2000, 323 cases of illness and one death were attributed to consumption of processed meats with a total cost to the community estimated to be $77 million. Controlling pathogens in processed meats include effective cooking, curing or fermentation with starter culture, and implementation of good hygienic practices (GHP) to limit the potential for post-processing contamination with L. monocytogenes. Food safety programs have been widely implemented in the Food Safety Scheme Risk Assessment

Page 7 of 189

processed meat sector to ensure control measures are in place. However, it was estimated that if these programs are not complied with, and poorly controlled or unreliable processing was allowed to occur in the production of uncooked comminuted fermented meats, this could lead to a significant increase in risk, with estimates the number of foodborne illness cases in Australia due to pathogenic E. coli could be up to 604 cases per year.

Plant products food safety scheme
Previous risk assessment work conducted on the risk associated with plant products found that fresh cut fruit and vegetables, seed spouts, vegetables in oil and unpasteurised juice present a high risk. This was due to a history of foodborne illness outbreaks in Australia, mainly due to Salmonella, in addition to a number of outbreaks overseas. Annual NSW consumption of these products has been estimated as being 11,000 tonnes of fresh cut vegetables, 150 tonnes of fresh cut fruits, 2600 tonnes of seed sprouts, 1000 tonnes of vegetables in oil products, and 100,000 L of unpasteurised juice.

Surveys of plant products have shown the potential for these high risk plant products to be contaminated with L. monocytogenes, Aeromonas spp., B. cereus and Salmonella.
Contamination of fresh cut fruit and vegetables can occur during growth, harvest or processing with the main pathogens of concern being L. monocytogenes in general, and C. botulinum for modified atmosphere packaged product. These products are considered high risk when they are consumed raw.

Seed sprouts can become contaminated with B. cereus, Salmonella and pathogenic E. coli during growth and harvest of the seeds and also during the sprouting process, which provides a near perfect environment for the growth of microorganisms. The oxygen reduced environment provided by vegetables immersed in oil allows for the growth of anaerobic microorganisms including C. botulinum, the cause of botulism. To reduce the risk, the
vegetables or fruits are usually cooked and acidified prior to placement in oil.

Unpasteurised fruit juices may become contaminated during the juicing process, either due to contamination on the exterior of the fruit or the use of damaged and mouldy fruit. Because the juice is not heat treated, any pathogenic microorganisms present are able to survive, and acid tolerant strains of pathogenic E. coli and Salmonella may grow.

Seafood safety scheme
Annual consumption of seafood has been estimated at approximately 15.1 kg per person. The hazards associated with seafood vary depending on the type of seafood and processing methods employed.
Shellfish, particularly oysters, as filter feeders can accumulate contaminants from the growing environment. The hazards of concern for shellfish include pathogenic bacteria and viruses, algal toxins and chemical contaminant from the growing environment. Viral contamination of shellfish is recognised as the highest risk for all seafood and is effectively managed by the implementation of shellfish safety programs that manage the waterway and harvesting of the shellfish. Algal toxins in shellfish are generally considered low risk where harvest management programs manage the risk. Where no programs are in place, the risk associated with algal

Food Safety Scheme Risk Assessment

Page 8 of 189

toxins increases to medium. Severe illness has been associated with the consumption of oysters contaminated with V. vulnificus.
Wild caught prawns particularly those caught in estuarine waters are susceptible to contamination from the environment, such as naturally occurring Vibrio spp. present in the waterway. Prawns treated with metabisulphite may present a problem to consumers who suffer from asthma-related conditions. The on-board cooking and cooling of prawns also has the potential to introduce bacterial contamination when water from the
waterway is used to cool the prawns.

Hazards associated with wild caught finfish include ciguatera and histamine, depending on the type of fish caught. Ciguatera poisoning is generally regarded as medium risk, with most illnesses occurring with fish caught near tropical reefs. Histamine poisoning is usually associated with certain fish such as tuna, swordfish, mahi mahi and blue grenadier and can be controlled by effective temperature control throughout the supply chain. If the fish are to be consumed raw, hazards such as parasites and Vibrio spp. become significant. Mercury in finfish presents a risk to pregnant women or women planning to become pregnant. Because mercury is naturally present in the marine environment, management strategies have relied on education of the consumer, in particular advising pregnant women to avoid consuming large predatory fish which are known to contain higher levels of mercury. Processed, RTE seafood products (including smoked seafood) can support the growth of L. monocytogenes, however contamination is thought to occur during the handling and packaging of the finished product. Strict hygiene and sanitation programs can reduce the likelihood of contamination. The packaging of smoked seafood under modified atmosphere packaging may allow the growth of C. botulinum. While botulism poisoning associated these products is rare, the illness is severe and is considered a medium risk.

Vulnerable persons food safety scheme
Certain population sub-groups are more at risk of foodborne illness or can develop more severe conditions due to foodborne illness when compared to the general population. The degree of vulnerability depends on the susceptibility of the individual and the pathogenicity of the pathogenic microorganism. In general terms, the vulnerable population group includes children under five years of age, people over 65 years old, pregnant women and persons with depressed immunity. It is estimated that the number of meals served to vulnerable persons in NSW facilities such as hospitals, aged-care facilities, hospices, day care establishments and childcare centres is approximately 133 million meals per year. It is estimated that up to one million meals per year served at these facilities may be contaminated with a foodborne pathogen.

Since 1995 there have been 65 foodborne illness outbreaks in Australian aged-care facilities, childcare centres and hospitals with 758 illnesses and 75 fatalities. The pathogens implicated have included Salmonella, C. perfringens, L. monocytogenes and Campylobacter. The prevalence of foodborne-related illness and deaths in the elderly living in nursing homes is far greater than the baseline level of illness in general population, while children appear more at risk to Salmonella due high salmonellosis rate in children seen both in Australia and overseas. The major hazard of concern to vulnerable persons is L. monocytogenes, with some sub-groups within the vulnerable population being 100 times more susceptible to listeriosis than the general population. Other hazards of concern include infants Food Safety Scheme Risk Assessment

Page 9 of 189

exposed to C. botulinum through consumption of contaminated honey, neonatal infants consuming Cronobacter sakazakii (formerly Enterobacter sakazakii) contaminated infant formula and individuals with liver dysfunction exposed to Vibrio vulnificus via raw oysters. Other organisms that may result in more severe illness in vulnerable sub-groups include pathogenic Enterohaemorragic E. coli, S. aureus and C. perfringens.

When assessing the risk associated with foods, it is important to consider food preparation and hazardous scenarios. Businesses catering to vulnerable persons need to consider the susceptibility of their consumers when designing menus and sourcing, preparing and serving foods. Under the Food Standards Code, these establishments are required to implement a food safety program including, substitution of high risk foods with lower alternatives; effective cleaning and sanitation of fruits and vegetables to be consumed raw; limiting the storage of reconstituted infant formula; minimising storage times/temperatures for RTE foods; ensuring foods are cooked properly and effective cleaning and sanitation of equipment.

Egg food safety scheme (draft)

The average Australian consumes approximately 137 eggs per year, equating to over 800 million eggs being consumed in NSW each year. The primary hazard of concern is Salmonella, in particular Salmonella Typhimurium which may contaminate the egg shell through environmental contamination and through contact with bird faeces. Overseas foodborne illness outbreaks attributed to eggs have been predominantly due to Salmonella Enteritidis however, Australian layer flocks remain free of Salmonella Enteritidis.

While Salmonella may be present in the farm environment, surveys have found the prevalence of Salmonella on shell eggs to be very low. However, eggs and egg products can also become contaminated during the grading and processing due to improper crack detection, incorrect washing of eggs and poor hygiene and sanitation during the processing of eggs into pulp and other products. Although egg products such as liquid pulp are pasteurised, the heat treatment is only mild and therefore it is important to limit the level of microbiological contamination. There is significant epidemiological evidence to suggest that a major contributing factor of salmonellosis in Australia is the use of dirty and cracked eggs, especially in products that receive minimal or no cook step. The Food Standards Code limits the sale of cracked eggs to businesses where the egg will be further processed and receive a heat treatment.

Depending on the hygienic practices on farm and proper grading, processing and storage of eggs, the potential number of egg-related illnesses was estimated up to 1800 cases per year across Australia. Current industry practices to address these issues include strict biosecurity on farm, implementation of quality assurance systems during grading and processing and effective supply chain management. Potential sources of chemical contamination of eggs on farm include contaminated soil, insecticide spray, incorrect use of medication and inappropriate egg washing solutions and concentrations. However, in general, only low levels of chemicals have been detected in eggs and previous risk assessment has assessed the risk of chemicals in eggs as low. The exception to this are specialty eggs such as Balut, salted and century eggs, where surveys have detected the unauthorised use of lead as an additive, leading to chemical contamination of some
products. These products Food Safety Scheme Risk Assessment

Page 10 of 189

may also become contaminated with pathogens due to the extensive handling during processing.

Conclusion
This review has illustrated that across the food safety schemes there are many potential hazards that can impact on human health with microbiological hazards considered the most significant. It concludes that for food businesses within these schemes, mitigating food safety risks requires the development and implementation of reliable, systematic and preventative procedures. Such procedures are the core elements of food safety programs, introduced either due to regulatory requirements or through industry-sponsored codes of practice. The review acknowledges that mitigating food safety risk necessitates a multi-factorial approach extending beyond the controls implemented by a food business operating under a food safety scheme.

Food Safety Scheme Risk Assessment

Page 11 of 189

Introduction
Purpose
Under current legislation, the NSW Food Authority (the Authority) can establish food safety schemes in respect to different type or classes of foods, food businesses or food activity (Food Act 2003). The food safety schemes aims to assist in improving the safe production and handling food by outlining the regulatory requirements for the food, food businesses or food activity (or activities) covered by the scheme. Currently food safety schemes exist for dairy, meat, plant products and seafood businesses operating in NSW, as well as businesses serving food to vulnerable persons. In addition to these schemes that have already been implemented, an egg food
safety scheme is currently being finalised.

These commodities have been identified as containing high risk products that may potentially cause outbreaks of foodborne illness, and where the cost benefit analysis justified a regulatory presence, and the use of regulatory tools such as the implementation of food safety programs based on principles of Hazards Analysis Critical Control Point (HACCP).

Current legislation also requires a risk assessment be undertaken when establishing a new food safety scheme. The risk assessment provides the science to underpin the food safety scheme and is required to be based on national or international standards.

The Authority, or its predecessor SafeFood Production NSW, previously commissioned risk assessment prior to the introduction of food safety schemes relating to seafood and plant products. In addition, risk assessments were conducted on the proposed scheme for egg and egg products and during the review of the dairy food safety scheme. The requirements under the meat food safety scheme were carried over from a previous legislation which did not require a risk assessment to be conducted and as such the Authority has not previously conducted a risk assessment in relation to meat. The vulnerable persons food safety scheme was introduced following the gazettal of Standard 3.3.1 – Food Safety Programs for Food Service to Vulnerable Populations of the Australia New Zealand Food Standards Code. Food service to vulnerable populations were identified as high risk in the National Risk Validation Project (Food Science Australia and Minter Ellison, 2002) The purpose of this risk assessment document is to provide a scientific review of hazards and their associated risks for food businesses covered by the food safety schemes. This risk assessment summarises the known information from previous risk assessment and where new or updated information is available, this has been incorporated into the information.

Scope
This risk assessment will review the hazards associated with food businesses
regulated under the food safety schemes of the NSW Food Regulation 2004 and includes:
•

Dairy

•

Meat

•

Plant products – fresh cut fruits and vegetables, unpasteurised juice and vegetables in oil

Food Safety Scheme Risk Assessment

Page 12 of 189

•

Seafood

•

Food service to vulnerable persons

•

Eggs and egg products (draft food safety scheme currently being finalised)

Overview of risk assessment
Risk assessment forms part of an overall process, called risk analysis. Risk analysis is used by governments and industry to assess, manage and communicate the risk associated with particular food or food groups and in
turn aims to reduce the potential for foodborne illness. The Codex Alimentarius Commission divides risk analysis into three components (CAC, 1999):

•

Risk assessment – a process by which the potential risk posed by food safety hazard(s) is determined

•

Risk management – the process of determining alternatives for control the hazards identified in the risk assessment and

•

Risk communication – the exchange of information on risk and risk management amongst interested parties.

CAC (1999) has identified four components of risk assessment: •

Hazard identification – the process of identifying potential hazards associated with the food

•

Exposure assessment – an estimation of the potential human exposure to the hazard and includes the use of data such as the occurrence in the food and/or potential consumption rates of the food

•

Hazard characterisation – the evaluation of the potential illness associated with the hazard

•

Risk characterisation – the process of determining the probability of occurrence and severity of the adverse health effects based on the information collected in the hazard identification, exposure assessment and hazard characterisation.

Previous risk assessment work
When developing a food safety scheme, the NSW Food Authority has previously commissioned hazard assessments, risk profiles and risk assessments or sourced information from risk assessments performed by other government and nongovernment organisations. These risk assessments have included: •

Ross, T. & Sanderson, K. (1999). A risk assessment of selected seafoods in NSW

•

Food Science Australia (2000). Final report – Scoping study on the risk of plant products

•

Food Science Australia & Minter Ellison Consulting (2002). National risk validation project. Final report

•

Miles, D. (2004). Risk assessment of the NSW dairy industry (unpublished)

•

Miles, D. and Chan, C. (2007). Risk profile and risk management of eggs and egg products in NSW (unpublished)

Food Safety Scheme Risk Assessment

Page 13 of 189

In addition, risk assessments have been conducted in Australia by other government and non-government organisations. These include:
•

Sumner, J. (2002). Food safety risk profile for primary industries in South Australia. Department of Primary Industries and Resources SA, Adelaide

•

FSANZ [Food Standards Australia New Zealand] (2002). Final assessment report proposal P263. Safety assessment of raw milk very hard cooked-curd cheeses. Food Standards Australia New Zealand Report

•

FSANZ (2006). A risk profile of dairy products in Australia. Food Standards Australia New Zealand Report

•

MLA [Meat & Livestock Australia] (2003). Through chain risk profile for the Australian red meat industry Part 1: Risk Profile

•

FSANZ (2005). Scientific assessment of the public health and safety of poultry meat in Australia

•

FSANZ (2006). Public health and safety of poultry meat in Australia – Explanatory summary of the scientific assessment, Canberra

•

MLA (2006) Listeria monocytogenes in smallgoods: Risks and controls

•

Ross, T. Walsh, P. and Lewis, T. (2002). Risk assessment of fish cold smoking and marination processes used by Australian businesses. Biodevelopment Consulting Pty. Ltd for SafeFood Production NSW

•

FSANZ (2005). Final assessment report, P265, primary production and processing standard for seafood (Attachment 10)

•

Daughtry, B., Sumner, J. Hooper, G., Thomas, C. Grime, T., Horn, R., Moses, A. & Pointon, A. (2005). National food safety risk profile of egg and egg products. A report for the Australian Egg Corporation Limited (AECL) Publication No 05/06 Project SAR-47

•

Thomas, C., Daughtry, B., Padula, D., Jordan, D., Arzey, G., Davey, K., Holds, G., Slade, J., & Pointon, A. (2006). An egg: Salmonella quantitative risk assessment model. AECL publication

Current approach
As there has been a considerable amount of risk assessment work already undertaken on industries covered by the food safety schemes, the approach taken in this document was to provide a review of previous work conducted. This information has been supplemented with other more recently published information where necessary (CAC, 2007). To minimise repetition, information
common to the different food safety schemes has been placed in the appendices to the document: •

Appendix 1 – Common microbiological and chemical food safety hazards

•

Appendix 2 – Food recalls in Australia

•

Appendix 3 – Foodborne illness outbreaks in Australia

A number of methods have been used to estimate the level of exposure to hazards. The approaches used include production data, consumption data, imported foods testing failures, recalls, epidemiological data and results of food surveys. When considering exposure the fate of the hazard during processing and preparation must be taken into account.

Food Safety Scheme Risk Assessment

Page 14 of 189

References
CAC [Codex Alimentarius Commission] (1999). Principles and guidelines for the conduct of microbiological risk assessment. CAC/GL-30. Retrieved 14 October 2008, from . CAC [Codex Alimentarius Commission] (2007). Working principles for risk analysis for food safety for application by governments. CAC/GL 62/2007. Retrieved 22 December 2008, from http://www.codexalimentarius.net/download/standards/10751/CXG_062e.pdf.

Food Act 2003, New South Wales Government (2008).
Food Regulation 2004, New South Wales Government (2008).

Food Science Australia & Minter Ellison Consulting (2002). National risk validation project. Final report 2002.

Food Safety Scheme Risk Assessment

Page 15 of 189

Dairy food safety scheme
Hazard identification
The safety of milk and milk products has been extensively reviewed by regulatory agencies in Australia and internationally. A large number of risk assessments and risk profiles have been undertaken, examining the risks across the entire dairy supply chain and conducting in-depth evaluations of specific pathogen-product combinations. This risk assessment will summarise the major body of relevant work undertaken to date.

In 1999, the former NSW Dairy Corporation commissioned Food Science Australia to review the food safety systems that had been implemented in the NSW dairy industry (Jansson et al, 1999). The report included a brief risk assessment and endorsed the preventative approach to food safety through the implementation of HACCP-based food safety programs throughout the dairy supply chain. This was later updated in 2004, when the NSW Food Authority completed a qualitative risk assessment of the NSW dairy industry which examined the microbiological and chemical hazards along the dairy supply chain (Miles, 2004). This was developed as an internal document to provide scientific rigour to the updated dairy food safety scheme and provide a basis for determining the priority classification for segments of the industry. In 2002, Primary Industries and Resources South Australia (PIRSA) commissioned a food safety risk profile on primary production, including milk and dairy products (Sumner, 2002). This report highlighted consumption of raw (unpasteurised) milk as a high risk activity.

In 2006, Food Standards Australia New Zealand (FSANZ) undertook a comprehensive risk profile of dairy products in Australia to inform the development of the Primary Production and Processing Standard for dairy
products (FSANZ, 2006). The FSANZ risk profile examined both microbiological and chemical hazards. The findings of the FSANZ risk profile are reported here.

Microbiological hazards
A wide range of microbiological hazards may be introduced into milk during primary production and processing. Raw milk may have a diverse range of bacteria present in it, either shed directly into the milk from the udder as a result of illness or disease, or through contamination from the external surface of the cow and the milking environment. FSANZ (2006) highlighted the on-farm factors that most significantly impact on the microbiological quality of raw milk as:

•

animal-related factors (eg animal health, herd size, age and production status)

•

environmental factors (eg housing, faeces, feed, soil, and water)

•

method of milking, operation of milking and storage equipment (eg cleanliness of equipment and lines, appropriate storage temperature to limit pathogen growth)

The initial levels of bacteria in raw milk can vary considerably, dependent on the level of control over these factors. Boor (1997) reviewed the different types of pathogenic microorganisms that have been detected in raw milk (Table 1).

Food Safety Scheme Risk Assessment

Page 16 of 189

Table 1 – Pathogenic microorganisms detected in raw milk
Microorganism
Enterobacteriaceae
pathogenic Escherichia coli (eg EHEC, STEC)

Disease

Salmonella
Shigella
Yersinia enterocolitica
Cronobacter sakazakii

Gastroenteritis, other complications
involve Haemolytic uraemic
syndrome (HUS) and Thrombotic
thrombocytopenic purpura (TTP)
Gastroenteritis, typhoid fever
Dysentery
Gastroenteritis
Meningitis in premature infants

Campylobacter jejuni
Aeromonas hydrophila

Gastroenteritis
Gastroenteritis

Pseudomonas aeruginosa
Brucella spp.

Gastroenteritis
Brucellosis (Bang’s Disease)

Bacillus cereus
Bacillus anthracis
Clostridium perfringens
Clostridium botulinum

Gastroenteritis
Anthrax
Gastroenteritis
Botulism

Staphylococcus aureus
Streptococcus agalactiae
Streptococcus pyogenes
Streptococcus zooepidemicus

Emetic intoxication
Sore throat
Scarlet fever/sore throat
Pharyngitis, nephritic sequelae

Listeria monocytogenes
Corynebacterium spp.
Mycobacterium bovis
Mycobacterium tuberculosis
Mycobacterium paratuberculosis

Listeriosis (various manifestations)
Diphtheria
Tuberculosis
Tuberculosis
Johne’s disease (ruminants)
Crohn’s disease (unproven in
humans)

Vibrionaceae and Campylobacter

Other Gram-negatives

Gram-positive sporeformers

Gram-positive cocci

Miscellaneous Gram-positives

Rickettsia

Coxiella burnetii

Viral
Enteroviruses, including polioviruses and Coxsackie
virus, Rotaviruses
Foot and mouth disease virus
Hepatitis virus
Fungi
Mould (and associated aflatoxins)
Protozoan parasites

Cryptosporidium parvum
Entamoeba histolytica
Giardia lamblia
Toxoplasma gondii

Q fever
Enteric infection
Foot-and-mouth disease
(not a human disease)
Infectious hepatitis
Mycotoxicoses
Cryptosporidiosis
Amoebiasis

Giardiasis
Toxoplasmosis

adapted from Boor (1997)

Food Safety Scheme Risk Assessment

Page 17 of 189

In the past, prior to the introduction of mandatory pasteurisation for milk in Australia, the most important human diseases disseminated by the consumption of raw milk were tuberculosis and brucellosis. These diseases have now been eradicated in Australian dairy cow herds. As a result, the FSANZ risk profile (FSANZ, 2006) went on to identify the most significant pathogenic microorganisms to public health and safety for the Australian dairy industry (Table 2). Further details on these microbiological hazards are available in Appendix 1.

Table 2 – Microbiological hazards in dairy products
Pathogens
pathogenic Escherichia coli

Significance in dairy products
Pathogenic strains of E. coli can be found in cattle and may enter milk through faecal contamination. Is destroyed by pasteurisation

Salmonella

Salmonella is occasionally present in raw milk but is destroyed by

Yersinia enterocolitica

Campylobacter spp.
Bacillus cereus

Clostridium botulinum
Staphylococcus aureus
Listeria monocytogenes
Cronobacter sakazakii
(formerly Enterobacter
sakazakii)

pasteurisation. Can contaminate products after pasteurisation, with nondairy ingredients a source of contamination. Frequently isolated in milk powder
Y. enterocolitica is destroyed by pasteurisation and its presence in heat treated milk products is due to environmental contamination after heat treatment. Y. enterocolitica is able to grow in dairy products held at refrigeration temperatures and therefore may be considered as a hazard in prolonged shelf life products

Campylobacter spp. is destroyed by pasteurisation and its presence in milk products is due to environmental contamination after heat treatment. Not normally able to grow in foods
Vegetative cells of B. cereus do not survive pasteurisation, however spores will survive. B. cereus is rapidly outgrown by psychrotrophic bacteria at refrigeration temperatures. However, in the absence of a competitive microflora, growth to levels of concern is possible Vegetative cells of C. botulinum do not survive pasteurisation, however spores will survive. Will only grow under anaerobic conditions May enter raw milk through udder infection. S. aureus is destroyed by pasteurisation, however toxins are heat stable. S. aureus does not grow well at refrigeration temperatures or compete with starter cultures L. monocytogenes is destroyed by pasteurisation. Its presence in dairy products is due to post-pasteurisation contamination. Can grow in milk products at refrigeration temperatures

C. sakazakii will not survive pasteurisation. Recontamination of powdered infant formula during manufacture is a risk. C. sakazakii cannot grow in a dry substrate, but it can survive for long periods of time and is a potential hazard when the powder is reconstituted and held at favourable temperatures. Contamination and subsequent growth may occur during
reconstitution and preparation

adapted from FSANZ (2006)

These hazards were considered significant due to either:
•

association with reported incidents of foodborne illness (including overseas outbreaks), or

•

the potential to contaminate dairy products after pasteurisation.

This conclusion is supported by other work, such as that by Todd & Harwig (1996) in a risk analysis of Canadian food, who stated that although 21 microbial hazards have been reported to occur in Canadian milk, only eight of those presented a significant risk to the human population, with Campylobacter jejuni, Salmonella serovars and E. coli O157:H7 identified as the most important hazards. Johnson et al (1990) identified Salmonella, Listeria monocytogenes and pathogenic E. coli as the three Food Safety Scheme Risk Assessment

Page 18 of 189

high risk organisms to the cheese industry in the USA. While, more recently, the presence of Cronobacter sakazakii in infant formula has presented a significant risk to premature infants (Lai, 2001; FDA, 2002; Himelright et al, 2002). Chemical hazards

Chemicals are used by the dairy industry for a number of purposes, including pest and weed control on farm, animal health and sanitising equipment. As a result, milk may be susceptible to chemical contamination if proper controls are not in place. The FSANZ risk profile evaluated the following potential chemical hazards (FSANZ, 2006): •

agricultural and veterinary chemical used in dairy primary production

•

environmental contaminants, including heavy metals, organic contaminants and micro nutrients

•

naturally-occurring chemicals found in plants or in fungi or bacteria associated with plants which may be ingested by grazing cattle

•

food processing by-products

•

food additives, processing aids, and those chemicals that may migrate from packaging into dairy products

Dairy products must comply with Standard 1.4.1 – Contaminants and Natural Toxicants and Standard 1.4.2 – Maximum Residue Limits of the Food Standards Code. These Standards sets out the Maximum Levels (MLs) of specified metal, nonmetal contaminants and natural toxicants and the Maximum Residue Limits (MRLs) for agricultural and veterinary chemical residues present in food respectively.

Agricultural and veterinary chemical hazards
Without appropriate controls and the observance of appropriate withholding periods for treated dairy cattle, it is possible for residues of these chemicals to occur in raw milk. In Australia, the Australian Pesticide and Veterinary Medicines Authority (APVMA) is responsible for registering agricultural and veterinary chemical products, granting permits for use of
chemical products and regulating the sale of agricultural and veterinary chemical products. Veterinary chemicals administered to dairy cattle are mainly antimicrobials and endo- and ectoparasiticides. Other veterinary chemical uses include reproductive therapy and use of anti-inflammatory drugs or anaesthetics. If the cow is lactating, then the product must specifically state that it can be used in lactating dairy cows, and a milk withholding period may be specified. The use of environmentally persistent pesticides, such as organochlorines, still poses a potential problem for grazing animals. Potential hazards include excessive levels of herbicides, pesticides or fungicides. Cereals and treated seeds used as animal feed supplement are the most likely source of these contaminants, with the most significant hazard to human health being those chemicals that can accumulate in animal tissues or are excreted in the milk.

Aflatoxins
Grain crops can become contaminated with biological toxins, such as aflatoxins, a group of extremely toxic metabolites produced by the fungi Aspergillus flavus and Aspergillus parasiticus. When these moulds are allowed to germinate and grow on harvested seed crops, the aflatoxins can be formed and ingested by dairy cattle during feeding, eventually contaminating the milk. Aflatoxin contamination of milk is more common in Europe where intensive supplementary feeding of dairy herds is

Food Safety Scheme Risk Assessment

Page 19 of 189

conducted. In Australia, where herds predominantly graze on pasture, aflatoxin contamination has not been reported (ANZFA, 2001).

Cleaning chemicals
Milking premises and equipment must be cleaned and sanitised to prevent the risk of contaminating the milk with microbiological pathogens. However, overuse of these chemicals can in itself create a hazard with the risk of chemical residues being left on equipment. All chemicals used in detergents
and sanitisers have the potential to leave a residue on the dairy equipment surface if not used in the correct manner. Physical hazards

The probability of introduction on farm of physical hazards which end up in the final product is thought to be minimal. Any physical hazard contamination that may be introduced on farm should be removed at the farm level. Most dairy farms include a filter ‘sock’ through which the milk passes prior to entering into the farm vat. This filter will remove most gross physical contaminants.

The introduction of physical hazards at the processing level has occasionally happened in the past, with pieces of equipment ending up in a dairy product. The instances of this occurring are very rare, and the preventative maintenance of equipment means the risk is very low. Good manufacturing practices and staff training should also ensure that the risk of physical contaminants through the wearing of personal effects such as jewellery are minimal.

Exposure assessment
Consumption of pasteurised milk and dairy products
Consumption of milk and milk products forms a significant part of the average Australian’s diet. Standard 4.2.4 – Primary Production and Processing Standard for Dairy Products of the Food Standards Code requires all milk for human consumption (including milk used to make dairy products) to be pasteurised at a minimum of 72°C for 15 seconds (or equivalent), unless an applicable law of a State or Territory provides an exemption 1. There is no such exemption for cows milk in NSW, therefore all dairy products for human consumption commercially sold in NSW are made from pasteurised milk. In 2006/07, 684 million litres of milk were sold in NSW/ACT, including modified and flavoured milk. Data from Dairy Australia shows the average consumption of dairy products in Australia each year is 103.6 L milk, 11.9 kg cheese, 6.8 kg yoghurt and 3.9 kg butter/blends per person (Dairy Australia, 2007). A closer analysis of consumption trends is shown in the Australian National Nutrition Survey (ABS, 1995), which showed that 84% of people surveyed consumed dairy products at a median amount of 347 g/day with the quantity
ranging from 209–471 g/day (Table 3). Consumption of dairy products varies with age, declining from 98% for children aged 2–3 years to 90% for adults aged 19–24 years and then increasing again to 95% for persons aged 65 years and over.

1

Standard 4.2.4A – Primary Production and Processing Standard for Specific Cheeses of the Food Standards Code does allow imported Gruyere, Sbrinz, Emmental and Roquefort to be made from raw milk.

Food Safety Scheme Risk Assessment

Page 20 of 189

Table 3 – Consumption of dairy products
Sex

Age

Male
Male
Male
Male
Male
Male
Male
Male
Male
Female
Female
Female
Female
Female
Female

Female
Female
Female

Proportion of persons
consuming milk products
and dishes 2
(%)

2–3
4–7
8 – 11
12 – 15
16 – 18
19 – 24
25 – 44
45 – 64
65+
2–3
4–7
8 – 11
12 – 15
16 – 18
19 – 24
25 – 44
45 – 64
65+

98.2
95.5
90.9
92.8
94.2
89.1
93.7

91.3
94.5
98.1
96.0
93.3
90.8
87.3
90.1
94.3
94.7
95.6

Median daily intake per
consumer of milk products and
dishes
(g/day)
471.8
365.0
401.4
424.0
392.2
323.0
263.2
258.0
255.0
394.3
280.5
312.0
297.7
258.0
251.3
209.3
216.6
225.8

adapted from National Nutrition Survey (ABS, 1995)

Liquid milk accounts for approximately 70% of the mean daily intake of dairy products for persons of all ages. However, the trend of milk consumption within Australia has been changing to more specialty types. Whole milk accounts for around 56% of milk sales, with lower fat lines increasing to 26%, long life or ultra high temperature (UHT) treated milk 8.5%, and the remainder as flavoured and specialty milks (Dairy Australia, 2007). Cheese consumption in Australia has jumped more than 20% in the past decade. The recent consumer trend has been away from cheddar cheeses to non-cheddar cheese types, and this is also being reflected in Australia’s cheese exports, where the non-cheddar share of total export sales has increased from 45% to 57% over the past seven years.

Consumption of raw milk
The current Dairy food safety scheme in the Food Regulation 2004 does not provide an exemption from the requirement to pasteurise cows milk sold for human consumption. However, no such limit exists on the private consumption of raw cows milk. This is believed to be limited to small communities in NSW, such as farm families. The amount of raw milk consumed on farm within NSW is difficult to estimate, but is considered to be extremely small when compared to the volume of pasteurised milk. The US Food and Drug Administration (FDA) and US Department of 2

Milk products and dishes are defined in the National Nutrition Survey (ABS, 1995) as including the following: – Dairy milk
– Yoghurt
– Cream
– Cheese
– Frozen milk products (eg ice -cream)
– Other dishes where milk or a milk product is the major component – Milk substitutes (eg soy-based milk)
– Flavoured milks

Food Safety Scheme Risk Assessment

Page 21 of 189

Agriculture (USDA) in its quantitative risk assessment on listeriosis estimated raw milk consumption to be less than 0.5% of total milk consumption in the USA (FDA/USDA, 2003). Todd & Harwig (1996) made the assumption that ‘farm families’ were the people most likely to consume unpasteurised dairy products and consequently be exposed to the microbiological hazards that may contaminate raw milk.

The dairy food safety scheme does provide an exemption to allow the sale of unpasteurised goats milk in NSW. This was initially a continuation of the permit system implemented by NSW Health. The former SafeFood NSW commissioned a risk assessment (AgriQuality New Zealand, 2002), but the authors could not fully determine the risk from unpasteurised goats milk due to a lack of data. Recommendations from the risk assessment report included the implementation of HACCP-based food safety programs and a microbiological survey of unpasteurised goats milk to generate data to provide the basis for a risk assessment. The National Nutrition Survey (ABS, 1995) estimated that less than 1% of respondents consumed goats milk and there is no evidence to suggest this has increased in recent years. In fact the number of licensed goat milk producers in NSW has declined. Recently ‘cosmetic’ and ‘bathing’ raw milk products have become available for sale in NSW and other states. Although marketed for non-food use, it is believed these labelling terms are being used to bypass the Food Standards Code, and that these products are being consumed. While the volume consumed is considered to be very small, the products are potentially unsafe. As such, the NSW Food Authority has taken enforcement action when these products have been identified in the marketplace, as they do not comply with the Food Standards Code requirements for pasteurisation of cows milk for human consumption.

FSANZ are currently considering Proposal P1007 – Primary Production & Processing Requirements for raw milk products. The outcome of that process could influence the volumes of unpasteurised dairy product offered for sale in Australia.

Hazard characterisation
Foodborne illness outbreaks from milk and dairy products
Australian dairy products enjoy a reputation for high standards of quality and safety. There have been few reported failures leading to incidents of foodborne illness attributed to dairy products in the market place in recent years. FSANZ reviewed the foodborne illness data associated with milk and milk products in Australia (FSANZ, 2006). This data is summarised in Table 4, with more detailed information on each outbreak included in Table 65 (Appendix 3). Between 1995 and 2008 there were 14 reported outbreaks directly attributed to specific dairy products, affecting 284 people. Of these, nine were associated with consumption of unpasteurised milk and none occurred in NSW. In addition, there were 12 outbreaks identified involving a food product that contained dairy products as an ingredient. However, because dairy products are an ingredient in many foods, it is often difficult to determine whether they are the actual cause of an outbreak.

There have been a number of reports of outbreaks associated with consumption of dairy products overseas. While unpasteurised dairy products have been a common cause of dairy-associated outbreaks of illness, pasteurised dairy products have also been implicated where there have been poor food safety control measures in place, including the use of contaminated non-dairy ingredients, faulty pasteurisation process, poor hygiene or contamination post pasteurisation.

Food Safety Scheme Risk Assessment

Page 22 of 189

Table 4 – Summary of foodborne illness outbreaks attributed to dairy products and foods including dairy as an ingredient
Hazard

Salmonella serovars
Campylobacter

Norovirus

C. perfringens
Chemical contamination

Cryptosporidium
S. aureus
Unknown
Total

Australian
outbreaks
(1995–2008)
7
6
3
1
1
1
1
6
26

Cases
226
85
123
27
23
8
2
86
582

Hospitalisations Deaths

5
0
0
0
0
3
0
1
10

0
0
0
0
0
0
0
0
0

adapted from FSANZ (2006)

While there is little corresponding evidence in Australia linking consumption of pasteurised dairy products to foodborne illness, the potential impact of a dairyrelated food poisoning outbreak due to the widespread consumption of dairy foods has been demonstrated by some large scale foodborne illness outbreaks overseas. In 2000, over 13,000 people were sick and 165 people hospitalised in Japan following consumption of dairy products made by the Snow Brand Milk Products Co. that were contaminated with S. aureus enterotoxin (Asao, 2003). In addition, the 2008 deliberate adulteration of Chinese milk with melamine resulted in worldwide recalls of dairy products and a wide range of other foods where milk powder was used as an ingredient. The broad distribution of Chinese milk powder graphically demonstrated that a food safety incident in one country can have international repercussions and encompass a broad spectrum of products.

Shiga toxigenic Escherichia coli (STEC) in raw milk
Cattle have been identified as an important reservoir for pathogenic E. coli, and although pasteurisation does eliminate E. coli, outbreaks of E. coli O157:H7 infections overseas have been attributed to contaminated raw milk and some pasteurised dairy products. A 1998 Australian study examined the incidence of STEC in dairy cattle on four farms, with evidence of STEC detected in the faeces of 39% of the 843 cattle tested (Desmarchelier, 1998). The prevalence rates varied between farms, although generally milking cows had a lower rate (24%) than younger animals (33–41%). The STEC isolated from dairy cattle included E. coli O157:H7 (0.9%) and E. coli O26 (0.16%), both known pathogenic serotypes. However, when the prevalence of STEC in Australian raw milk was assessed, it was found to be relatively low (Desmarchelier, 1998), with STEC isolated from 27 of 1,802 samples (1.5%). It was hypothesised that low level carriage may normally be present in the dairy herd and this is periodically stimulated by some host or environmental factor. It appears that during these episodes of increased faecal shedding, there is an increase in environmental contamination and associated increased risk of milk becoming contaminated.

Salm onella in dairy products
The first significant case of Salmonella in an Australian milk product occurred in 1943 in Victoria. A typhoid-carrying farm worker contaminated raw milk, which was then distributed for public consumption, resulting in over 400 cases of typhoid fever and Food Safety Scheme Risk Assessment

Page 23 of 189

23 deaths (Merrilees, 1943). Since that time, there has been only one other major incident involving Salmonella in pasteurised dairy products. This occurred in Victoria in 1977 and was traced to milk powder becoming contaminated due to contaminated lining of the spray dryer (Forsyth et al, 2003). The former Australian Dairy Authorities’ Standards Committee (now ANZDAC) produced the Australian manual for control of Salmonella in the dairy industry (ADASC, 1999b) to specify control measures for limiting the
risk of dried milk products becoming contaminated with Salmonella.

Yersinia enterocolitica in milk
Y. enterocolitica has been isolated from raw and pasteurised milk in various parts of the world. Although some strains occasionally associated with human disease have been isolated from raw milk, the main pathogenic types generally do not predominate. Milk has been associated with sporadic cases and outbreaks of Y. enterocolitica infections overseas where milk was contaminated postpasteurisation by contact with implements contaminated by milk crates used on piggeries (Barton & Robins-Brown, 2003). A suggested, though not proven, link between human yersiniosis and pasteurised milk in NSW has been reported (Butt et al, 1991).

Cam pylobacter spp. in milk
Both Campylobacter jejuni and Campylobacter coli are found in the faeces of cattle and can cause cases of subclinical mastitis. Hutchinson et al (1985) reported a milkborne outbreak resulting from the consumption of raw milk from cows exhibiting no outward evidence of illness. Healthy lactating cows can carry C. jejuni in the intestinal tract, providing an extrinsic source of contamination. In one US study of 193 healthy dairy cows at three dairies, 77 (40%) had positive rectal cultures (Martin et al, 1983).

Overseas surveys of Campylobacter in raw milk have shown a prevalence of 1 to 6% (Wallace, 2003). Campylobacter is killed by pasteurisation. However, these organisms are unlikely to grow in milk or dairy products. Nevertheless, several outbreaks of Campylobacter food poisoning from consumption of raw milk in Australia have been reported among children who were taken on a class trip to a dairy and given raw milk to drink.

Bacillus cereus in liquid milk
Raw milk is frequently contaminated with Bacillus spp. spores, with the milk often contaminated at the farm. Sanitation of the teats prior to milking was able to reduce the incidence of B. cereus in raw milk (Christiansson et al, 1999). The presence of B. cereus in a processed dairy product is often associated with the ability of the spores to survive pasteurisation, after
which the resulting vegetative cells may colonise pipes, tanks and filling machines (Lin et al, 1998). Notermans et al (1997) examined the risk from B. cereus in pasteurised liquid milk in the Netherlands. The study estimated that up to 7% of pasteurised milk may contain B. cereus, with levels ranging up to 105 cfu/mL.

In Australia, pasteurised milk has not figured as a cause of B. cereus food poisoning.

Clostridium botulinum in dairy products
Dairy products have not traditionally been associated with outbreaks of botulism. Since 1912 fewer than 12 outbreaks associated with dairy products worldwide have been recorded (Szabo & Gibson, 2003). However, spores of C. botulinum survive the Food Safety Scheme Risk Assessment

Page 24 of 189

normal milk pasteurisation process, and therefore control factors such as aw (water activity), redox potential, pH and temperature must be used in dairy products such as cheeses and dairy-based spreads and sauces to reduce the risk of botulism. In 2007, one case of botulism was reported in Victoria and was associated with the consumption of a nationally distributed ready-to-eat nachos meal. The neurotoxin was detected in discarded remains of that meal pathogen (OzFoodNet Working Group, 2007). One of the components was a cheese sauce and subsequent laboratory testing showed that the sauce provided an environment that would support growth of the organism. This case is not included in Appendix 3 data as it was not classed as an outbreak, affecting only one person.

Botulism in dairy herds is caused by ingestion of preformed toxins produced by the growth of C. botulinum in decaying crops, vegetation or carcase material, or by the animal acquiring a gastrointestinal infection with the organism. The presence of neurotoxin in milk from animals diagnosed with botulism is periodically raised as a concern. When these incidents occasionally occur in dairy herds in Australia, farmers voluntarily remove
affected animals from supplying milk. There is little evidence in the scientific literature to suggest the transfer of botulinum neurotoxins or the organism itself to milk occurs from either symptomatic or asymptomatic animals in affected herds.

Staphylococcus aureus in milk
S. aureus is a cause of mastitis in milk producing animals and can be frequently

found in raw milk from cows with undetected mastitis. Even in subclinical cases of mastitis up to 105 cfu/mL of S. aureus can be shed into the milk. S. aureus is a poor competitor, and will not grow well in the presence of other bacteria commonly present in raw milk. However, it is believed that toxin can be produced under any conditions that permit growth (Stewart, 2003). The Snow Brand Milk Products Co. outbreak in Japan was suspected to be due to poor cleaning of distribution pipes in the production facility, leading to the opportunity for S. aureus to grow to high levels and produce toxin (Asao, 2003).

Listeria m onocytogenes in dairy products
L. monocytogenes has a history of causing large outbreaks from dairy products, with a 1985 outbreak in the USA from Mexican-style soft cheese affecting 142 people and causing 48 deaths and an outbreak in Switzerland from Vacherin Mont D’Or cheese affecting 122 people and causing 34 deaths (Ryser, 1999).

The New Zealand Food Safety Authority (NZFSA) commissioned a series of risk profiles examining the risk of L. monocytogenes contamination in ice-cream, low moisture cheese and soft cheeses (see Table 5 for outcomes) while the FDA/USDA risk assessment on L. monocytogenes examined 11 categories of dairy products (FDA/USDA, 2003) (see Table 6 for outcomes extrapolated to the Australian population).

The worldwide incidence rate for Listeria spp. in raw milk is estimated to be around 3–4% (Sutherland et al, 2003), while in Australian raw milk the
incidence also appears low. A NSW Dairy Corporation survey of 600 raw milk samples failed to detect L. monocytogenes, however, 0.4% of samples were positive for Listeria spp. (Sutherland & Porritt, 1995).

The organism is eliminated by pasteurisation, therefore the primary concern is postpasteurisation contamination with L. monocytogenes , as it is a common inhabitant of dairy processing facilities. In dairy factories, the major areas that have been Food Safety Scheme Risk Assessment

Page 25 of 189

identified as sources of the organism are drains, floors, conveyors, refrigerated storage areas and crate wash lines (Sutherland et al, 2003).

Cronobacter sakazakii in infant formula (formerly Enterobacter sakazakii) C. sakazakii is a rare, but life threatening cause of neonatal meningitis, sepsis, and necrotising enterocolitis. In general, the reported case-fatality rate varies from 33– 80% among newborns diagnosed with this type of severe infection (Lai, 2001). Premature infants and those with underlying medical conditions may be at highest risk for developing a C. sakazakii infection. However, it should be noted that healthy infants may not always be immune to C. sakazakii infections (Nazarowec-White & Farber, 1997).

In 2002, the US FDA issued an alert to health care professionals regarding the risk associated with C. sakazakiiinfections among neonates fed milk-based, powdered infant formulas (FDA, 2002). There have been several C. sakazakiioutbreaks reported among infants fed milk-based powdered formula in neonatal intensive care units in England, Netherlands, Iceland, Belgium, Greece, U.S. and Canada (Biering et al, 1989; Lai, 2001; Van Acker et al, 2001; Himelright et al, 2002). These outbreaks have involved several deaths and were associated with temperature abuse of reconstituted powdered infant formula. In addition, there have been cases in premature babies in New Zealand (NZ Ministry of Health, 2005): •

1986 – a premature infant contracted C. sakazakii septicaemia. The infant survived, apparently without serious sequelae

•

1991 – premature twins contracted C. sakazakii meningitis. One twin suffered serious permanent neurological effects and the other recovered fully

•

2004 – a premature infant contracted C. sakazakii meningitis and died

At the time of writing, there have been no reported cases of neonatal illnesses associated with C. sakazakiiin infant formula in Australia. However it must be noted that the organism is not a notifiable disease in Australia.

Powdered infant formula is not a commercially sterile product, unlike liquid formula which is subjected to sufficient heat to render it commercially sterile. Powdered infant formula may be subject to contamination by opportunistic pathogens such as C. sakazakii through improper cleaning of production lines. While the pathogen does not grow in the powder it can survive for many months (Nazarowec-White & Farber, 1997)

Chemicals
The prevalence of chemical in dairy products is assessed by several surveys conducted each year in Australia to detect chemical residues. The Australian Milk Residue Analysis (AMRA) survey, the Australian Total Dietary Survey (ATDS), the National Antibacterial Residue Minimisation (NARM) program, and other targeted testing programs provide an indication of the potential for chemical contaminants ending up in dairy products.

The AMRA survey from 1998 to 2005 showed the following:
•

3,467 milk samples (89,121 analyses) for antimicrobials showed 99.997% compliance with the maximum residue limit (MRL) for veterinary chemicals residues in milk (there was one detection of Cloxacillin at a level at the MRL of 0.01 mg/kg in June 2002 in a bulk milk sample in NSW)

Food Safety Scheme Risk Assessment

Page 26 of 189

•

33,382 analyses for agricultural chemical residues, including organochlorines, organophosphates and synthetic pyrethroids, showed no detections

Targeted testing of milk in areas subject to locust plagues has also shown very high compliance rates for organochlorines, organophosphates and Fipronil (a broad spectrum insecticide). In 2000–2001, 123 samples were tested in NSW, with no residues detected (DFSV, 2002).

The NARM program conducts tests for antimicrobials on bobby calves and cull dairy cows that are presented to abattoirs. Testing on NSW cull dairy cows from 2000 to 2002 showed that 44 from 455 (9.7%) were positive. Of these positive trace results, eight were shown to be greater than MRL with Neomycin, and Sulphadiazine found in dairy cull cows, and Oxytetracycline and Sulphadiazine in the export calves. Chemical residues persist in meat much longer than milk, and this is reflected in recommended withholding periods. Traceback investigations where residues were detected showed that causes ranged from not obeying the withholding period, use of the wrong withholding period (milk rather than meat) and accidental feeding of medicated milk to calves (NSW Agriculture, 2001 NSW Agriculture, 2002).

The ATDS detected no agricultural chemical residues in milk and milk products available on retail shelves. Naturally occurring aflatoxins are not considered a high risk, as a small survey of 40 dairy products by the NSW
Food Authority in 2005 detected aflatoxin M1 in trace levels in only one sample, all others were below the limit of detection.

Risk characterisation
Risk ranking dairy products
The outcomes of the NZFSA risk profiles examining L. monocytogenes and STEC in dairy products are summarised in Table 5. The FDA/USDA (2003) estimated the risk per serving and risk per annum of listeriosis for eleven RTE dairy products, based on the predicted number of illnesses associated with the consumption of these foods. The risk was calculated for each food on both a ‘per serving’ and ‘per annum’ basis. Predictions based on the FDA/USDA (2003) risk assessment with a population of 260 million people were extrapolated to the Australian population of approximately 21.6 million (ABS, 2009) by dividing by a factor of twelve. The predicted number of annual listeriosis cases are presented in Table 6 noting that these estimates are approximate, as it is acknowledged that there may be differences in the consumption levels of particular dairy products between American and Australian consumers.

Food Safety Scheme Risk Assessment

Page 27 of 189

Table 5 – NZFSA risk profile outcomes examining hazards in dairy products Hazard

Listeria monocytogenes in ice-

cream
(Lake et al, 2003)

Listeria monocytogenes in low
moisture cheese
(Lake et al, 2005a)

Listeria monocytogenes in soft
cheeses
(Lake et al, 2005b)

Shiga-toxin producing
Escherichia coli in raw milk
(Gilbert et al, 2007)

Risk
Found no evidence to link consumption of ice-cream with
cases of L. monocytogenes infection in New Zealand
Contamination with L. monocytogenes is unlikely unless
introduced post-pasteurisation from environmental sources,
added ingredients or further processing such as grating.
Surveys of low moisture cheese suggest that contamination
with L. monocytogenes is infrequent and that growth in
product is unlikely. Even taking into account the high
consumption of low moisture cheese, the available data
indicates that L. monocytogenes in low moisture cheese
does not represent a significant risk to human health
Data on the prevalence of L. monocytogenes indicate that
contamination rates are very low, most likely to occur postpasteurisation. Current risk to the general population is considered low, although susceptible populations will have
a greater risk
Approximately 10% of notified human cases of STEC
infection (mostly E. coli O157:H7) in New Zealand reported
consumption of raw milk. E. coli O157 has been reported,
albeit rarely, in faecal samples from dairy and beef cattle. However, there is insufficient data on the prevalence and
numbers of STEC in raw milk to robustly estimate the risk
from consumption of raw milk in New Zealand

Dairy products that are likely to support the growth and survival of pathogens and are prone to contamination after pasteurisation may be
categorised as higher risk than other dairy products. Alternatively, dairy products that do not support the growth of pathogens, if correctly formulated, can be classified as low risk. However, it is also acknowledged that for some pathogens with a low infective dose, survival in the dairy product may become the issue more than the ability to grow. FSANZ ranked the degree of risk based on (FSANZ, 2006):

•

intrinsic properties of the product (ie the impact of aw, pH, salt concentration, and their effect on the growth of contaminating microorganisms)

•

extent to which food is exposed to factory environment or handling after heat treatment

•

hygiene and control during distribution and retail sale

•

degree of reheating or cooking before consumption (many dairy products are RTE, so this is rarely a factor)

Food Safety Scheme Risk Assessment

Page 28 of 189

Table 6 – Risk ranking for dairy products contaminated with Listeria m onocytogenes Dairy product

Risk ranking

(per serve)

Unpasteurised fluid milk
High fat and other dairy products (eg butter, cream,
other miscellaneous milk products)
Soft unripened cheese, >50% moisture (eg cottage
cheese, cream cheese, ricotta)
Pasteurised fluid milk
Fresh soft cheese
Semi-soft cheese, 39–50% moisture (blue, brick,
monterey, muenster)
Soft ripened cheese, >50% moisture (brie,
camembert, feta)
Ice-cream and other frozen dairy products
Processed cheese (cheese foods, spreads, slices)
Cultured milk products (yoghurt, sour cream,
buttermilk
Hard cheese, 6
11
23
2
>29
>50
5
50

Deaths
94
6
85
11
1
21
2
3

7
>7
11
3
20

adapted from MLA (2006); Doolittle (2008)

Table 15 – Prevalence of Listeria m onocytogenes in processed meats Product category
Processed meats (hams, whole muscle
cooked meats)
Cooked sausages (Frankfurters, saveloys)
Pâté and meat paste

Contamination rate (%)
4.8
2.8
1.2

adapted from MLA (2006)

Risk characterisation
Red meat
The MLA risk profile examined the risk of fresh meats consumed in the home and found that meat consumed as cuts, roasts, chops, steak are low risk (MLA, 2003a). These products normally receive a cooking step by the end consumer that will reliably eliminate most pathogenic bacteria. However, the importance of cross contamination from pathogens in raw and undercooked meat across to RTE foods was highlighted by Lake et al (2004) as a potential source of infection from Yersinia in pork (Table 16), and as a potential source for large numbers of Salmonella food poisoning cases (Table 17). It was predicted that if the cross contamination rate of Salmonella from raw meat to RTE foods increased from 1% to 10% it would result in almost an extra 4800 cases, while an increase to a 50% cross contamination rate would
result in more than 26,000 cases of foodborne illness in Australia each year, where fresh meat was the cause (MLA, 2003a).

Another hazard categorised as high risk by the MLA risk profile was undercooked sheep meat or liver contaminated with Toxoplasma gondii, particularly when Food Safety Scheme Risk Assessment

Page 52 of 189

consumed by pregnant women (MLA, 2003a). This area was identified as a data gap in work undertaken on the New Zealand meat industry by Lake et al (2002b). While it was predicted there may be more than 600 cases of illness due to infection withToxoplasma gondii per annum (200 in pregnant women), further work was required to fully understand the risk factors involved.

Undercooking of meat and hamburgers is associated with survival of pathogenic E. coli and the ranking of this pathogen as a medium hazard. However, a study by the FSIS (2002) in the USA calculated the probability of illness occurring as extremely low. This work estimated that the probability of E. coli O157:H7 surviving in a piece of cooked steak was 0.000026% (2.6 of every 10 million servings), given normal cooking practices. It was shown that even inadequate cooking of meat would still reduce the numbers of pathogenic E. coli present on the meat, albeit to a lesser degree than proper cooking. The predicted number of food poisoning cases from E. coli O157:H7 in cooked steaks was calculated to be one case for every 15.9 million servings (FSIS, 2002). The risk of illness from pathogenic E. coli in New Zealand meat was considered low by Lake et al (2002a), as there was no data to link illness to pathogenic E. coli in that country. The risk from comminuted meat such as hamburgers is considered greater, as the contamination may be spread throughout the product, as opposed to just the surface on an intact steak. The MLA risk profile estimated that if all hamburgers were appropriately cooked, there would be no illness. However, if 20% of hamburgers were undercooked, there would be six illnesses per annum in Australia (MLA, 2003a).

The NZFSA also commissioned risk profile work on tuberculosis (Cressey et al, 2006) and Campylobacter (Lake et al, 2007c) from red meat, and both were considered low risk (Table 16). These were not examined in the MLA risk profile, as Australia has been declared tuberculosis free and a definitive association of Campylobacter with red meat could not be established (MLA, 2003a).

The risk from C. perfringens from meat consumed in the home was considered medium (Table 17), with poor cooling and reheating the main contributing factors. Problems with C. perfringens have arisen more in large-scale, catering-type operations than in the home, where the spore forming bacteria have been allowed to germinate and grow due to poor cooling of meat-based dishes.

Food Safety Scheme Risk Assessment

Page 53 of 189

Table 16 – NZFSA risk profile outcomes examining hazards in meat Hazard

Risk

Shiga toxin-producing
Escherichia coli in red meat
and meat products

Overseas studies have consistently linked human cases of
STEC infection and particularly E. coli O157:H7 to
consumption of red meat in the form of undercooked
hamburgers, not one case in New Zealand has been
associated with regulated foods.

(Lake et al, 2002a)

Toxoplasma gondii in red meat

and meat products

(Lake et al, 2002b)

Yersinia enterocolitica in pork
(Lake et al, 2004)

Mycobacterium bovis in red
meat

(Cressey et al, 2006)

Campylobacter jejuni/coli in

red meat

(Lake et al, 2007c)

Food Safety Scheme Risk Assessment

Toxoplasma gondii is a protozoan parasite that causes

disease in humans with a range of outcomes including, at
worst, miscarriages. Cysts in the muscle tissue of meat
animals may result in infection when eaten. The
significance of human infections, especially congenital
toxoplasmosis, in New Zealand is unknown and has been
identified as a knowledge gap

Pigs are known to be frequently contaminated with
Y. enterocolitica, but effective cooking or pasteurisation will eliminate Y. enterocolitica from foods. Pork consumption
has consistently been associated with yersiniosis in studies in New Zealand and overseas, with cross contamination
from uncooked meats to RTE foods a potential source of

infection.
A proportion of human tuberculosis cases have been
caused by Mycobacterium bovis. While transmission of
tuberculosis to humans through consumption of M. bovisinfected meat is possible, no cases of this have been confirmed in New Zealand and it is considered low risk.
Seventeen outbreaks of campylobacteriosis in New Zealand
from 1999 to 2004 have been associated (weakly) with red
meat consumption. Data in New Zealand indicates there is
low but consistent contamination across pork, beef, and
sheep meat. On this basis it is identified as a minor risk
factor for exposure to Campylobacter in New Zealand.

Page 54 of 189

Table 17 – Risk ranking for meat and meat products
Meat product

Severity 7

Effect of
production,
processing,
handling on
the hazard
↓↑→
Retail meats consumed in the home (steak, mince, chops, roast, fresh sausages) Consumed
Toxoplasma
IB
Medium
No
↓freezing, →
undercooked/raw
gondii

Consumed
undercooked
Reheated roasts
Reheated roasts
Potential
pathogen
Poor cooling

7

8

Hazard

Probability

Growth
required
to cause
illness

Consumer
does
pathogen
reduction
step

Epidemiological
link

Risk
rating

Predicted
annual

number of
illnesses
(in
Australia) 8

Yes/No

Yes

High

EHEC

IB

Low

Yes

↓→

Yes

Yes

Medium

715 (242 in
pregnant
women)
No estimate

C. perfringens
S. aureus
Aeromonas

Mycobacterium
paratuberculosis
Bacillus
Yersinia
enterocolitica

III
III
III
III/IB

Low
Medium
Low
Low

Yes
Yes
Yes
N/A

↓↑
↓↑
↓↑
??

No
No
Yes
??

Yes
Yes
No
??

Medium
Low
Low
Low

No
No
No
No

III
III

Low?
Low??

Yes
Yes

↓↑
↓↑

Yes
Yes

??No
?

Low
Low

No estimate
No estimate

estimate
estimate
estimate
estimate

ICMSF (2002) defines the level of severity as follows:
− IA – Severe hazard for general population, life threatening or substantial chronic sequelae or long duration − IB – Severe hazard for restricted populations, life threatening or substantial chronic sequelae or long duration − II – High hazard incapacitating but not life threatening sequelae rare moderate duration. − III – Moderate, not usually life threatening no sequelae normally short duration symptoms are self limiting can be severe discomfort. Data from Sumner (2002) predicted annual numbers of illness per annum for the South Australia population (1.5 million), the MLA Risk Profile (MLA, 2003a) used an Australian population figure of 19.7 million. These estimates have been extrapolated to the current population of Australia estimated by ABS (2009) as approximately 21.6 million, by multiplying by a factor of 14.4 and 1.1 respectively.

Food Safety Scheme Risk Assessment

Page 55 of 189

Retail meats consumed in the home (steak, mince, chops, roast, fresh sausages) ↓↑
Assume 1% cross Salmonella
II/IB
Low
Yes
contamination
rate
↓↑
Assume 10%
Salmonella
II/IB

Low
Yes
cross
contamination
rate
↓↑
Assume 50%
Salmonella
II/IB
Low
Yes
cross
contamination
rate
Enterohaemorrhagic E. coli in hamburger
↓→
Hamburgers
EHEC
IA/IB
Low
No
↓→
Hamburgers EHEC
IA/IB
Low
No
assume 50%
undercooked
↓→
Hamburgers
Salmonella
II/IB
Low
Yes

Yes

Yes

Medium

583

Yes

Yes

High

5,830

Yes

Yes

High

29,370

Yes Yes

Yes
Yes

Low
Low

0
7

Yes

Yes

Low

0

adapted from Sumner (2002); MLA (2003a)

Food Safety Scheme Risk Assessment

Page 56 of 189

Table 18 – Risk ranking for processed poultry meat products Processed
meat
product

Hazard

Severity

Probability

Growth
required
to cause
illness

Consumer does
pathogen
reduction step

Epidemiological
link

Risk
rating

Predicted
annual
number of
illnesses (in
Australia) 9

Yes

Effect of
production,
processing,
handling on
the hazard
↓↑→
↓

Processed
chicken
Processed
chicken

Salmonella

II/IB

Low

No

Yes

High

864

Campylobacter

IB

Low

Yes

↓↑

No

Yes

High

86,400

adapted from Sumner (2002)

Table 19 – NZFSA risk profile outcomes examining hazards in poultry meat Hazard

Risk

Salmonella (non-typhoid) in poultry (whole and
pieces)
(Lake et al, 2004)
Campylobacter jejuni/coli in poultry
(Lake et al, 2007a)

Salmonellosis is the second most frequently notified enteric disease in New Zealand. Poultry meat is regarded as an important source of infection

Campylobacter jejuni/coli in mammalian and

poultry offals
(Lake et al, 2007b)

9

Campylobacter is the most frequently notified cause of enteric disease in New Zealand. Consumption of chicken was linked with Campylobacter infection and several outbreaks of campylobacteriosis identified undercooked chicken as the transmission vehicle

The consumption of poultry and mammalian offal is low in comparison to other meat types. However the high prevalence of Campylobacter in raw sheep and chicken livers is of concern, especially when some advice to consumers is to cook chicken livers “until they’re pink in the middle” or “lightly sautéed”. Offal for pet food is frequently contaminated and handling may provide a risk of infection

Data from Sumner (2002) predicted annual numbers of illness per annum for the South Australia population (1.5 million). These estimates have been extrapolated to the current population of Australia estimated by ABS (2009) as approximately 21.6 million, by multiplying by a factor of 14.4.

Food Safety Scheme Risk Assessment

Page 57 of 189

Poultry meat
In the FSANZ scientific assessment of poultry meat, it was found there was reasonable evidence to indicate poultry is the vehicle for a significant proportion of campylobacteriosis and salmonellosis cases in Australia
(FSANZ, 2006). This conclusion was made on the basis of epidemiological data, results from microbiological surveys of raw poultry carcase and outputs from a probabilistic model. Sumner (2002) estimated large numbers of foodborne illness cases to be due to processed chicken products, greater than 80,000 cases per annum from Salmonella and Campylobacter in Australia (Table 21). Work commissioned by the New Zealand Food Safety Authority has also found the presence of these organisms on poultry to be a significant source of infection (Table 19). Management of both Salmonella and Campylobacter requires an approach across both primary production and processing. Good hygienic practices and good agricultural practices are necessary prerequisites for the management of Salmonella and Campylobacter, and appropriate hygiene and sanitation is required during processing to minimise cross contamination between birds. Surveys of poultry pieces available for retail sale show that a large proportion of poultry carries these organisms, creating a risk of foodborne illness from consuming undercooked chicken, but also the risk of cross contamination occurring in food preparation areas. FSANZ (2005) found that the implementation of control measures to reduce the prevalence and levels of Salmonella and Campylobacter by ten-fold at the end of processing could result in a 74% and 93% reduction in the number of predicted cases of illness respectively.

FSANZ identified the following significant factors contributing to contamination of poultry meat with Salmonella and Campylobacter :
•

On-farm contamination with Salmonella is mainly due to contaminated feed and water, environmental sources and transmission from contaminated eggs

•

Important on-farm risk factors for Campylobacter are the age of the birds and environmental factors

•

The presence and amount of Salmonella on a chicken after processing largely determines the likelihood of salmonellosis

•

Inadequate hand washing and food handling practices determine the likelihood of human illness from Campylobacter

•

Adequate cooking is the main means of minimising the risk to human health from both pathogens

The FSANZ scientific assessment found little evidence of public health risks associated with chemical hazards from Australian poultry meat. The report concluded that the current regulatory measures appear to adequately protect public health and safety with respect to chemical hazards (FSANZ, 2006).

Game meat
Evidence suggests that the consumption of kangaroo meat and other game meat present little risk as a source of foodborne illness when compared to other forms of meat.

Food Safety Scheme Risk Assessment

Page 58 of 189

Processed meats
Sumner (2002) stated that the highest risk products in the meat industry in South Australia are smallgoods, predominantly due to pathogenic E. coli, Salmonella and L. monocytogenes. The risk of L. monocytogenes from processed RTE meats in New Zealand was examined by Lake et al (2002), who found that these products were a significant route of infection in that country. The FDA/USDA risk assessment for L. monocytogenes identified processed meats products or ‘deli meats’ as the highest risk food from the 23 RTE foods
examined in the USA (FDA/USDA, 2003). Deli meats were the only food to be ranked as very high risk and when extrapolated to the Australian population are predicted to be responsible for 133 cases of listeriosis per annum (see Table 22). In addition, frankfurters that were not reheated were ranked as high risk by the FDA/USDA in their risk assessment. However it is not believed that this practice is as common in Australia as in the US. Pâté and meat spreads were ranked as high risk on a per serving basis, predominantly due to the probability of contamination with L. monocytogenes post cooking, but this did not correlate with a high number of predicted illnesses due to low consumption rates. The MLA risk profile examined different scenarios for processed meat products and provided risk rankings and predicted numbers of illness (MLA, 2003a, summarised in Table 21). It was predicted that processed deli meats were responsible for 43 cases of listeriosis in Australia per annum. Although there are 50-60 cases of listeriosis reported each year in Australia, it is estimated that due to under-reporting, the true number is closer to 120 per year. Therefore, processed meats were considered to be the source of approximately one third of listeriosis cases in any one year (MLA, 2006).

The MLA guideline Listeria monocytogenes in smallgoods: risks and controls (MLA, 2006) proposes measures to control post processing contamination with L. monocytogenes in processed meat:
•

install effective GMPs and sanitation standard operating procedures (SSOPs), particularly in post-cooking operations such as slicing/portioning and packing

•

incorporate antimicrobials into formulations of products which are intended for slicing and packing as long shelf-life products

•

employ technologies for in-pack pasteurisation

In late 2008, the NSW Food Authority implemented national testing requirements for sliced pre-packaged RTE meat products (Meat Standards Committee, 2008). This Listeria management program involved the introduction of minimum finished product and environmental testing to any facilities manufacturing these products (NSW Food Authority, 2008).

The MLA risk profile (MLA, 2003a) and the work of Gilbert et al (2007) considered the risk of pathogenic E. coli (STEC and EHEC) from UCFM products such as salami. Both studies found that well controlled processes, including efficient fermentation and a maturation producing a low pH and water activity, effectively controlled the hazard (Table 20 and Table 21), with a 2-3 log reduction in E. coli. As a result, with a well controlled process, it was predicted that no illnesses would result from these products. However, if contaminated meat was used, and an unreliable process was not able to reduce the hazard, then there may be up to 604 illnesses in a year (see

Food Safety Scheme Risk Assessment

Page 59 of 189

Table 21). In addition, if susceptible members of the population consumed product with high levels of pathogenic E. coli present there may be up to 125 cases per annum. It was predicted that Salmonella may cause 11 illnesses per year from UCFM products, while green runners used for casings of sausages and salami would not cause any illness. Given the level of specialist skills and knowledge required to safely make these products, specific requirements for the production of UCFM products are included in Standard 4.2.3 – Primary Production and Processing Standard for Meat of the Food Standards Code. Salmonella was considered a medium risk, while the risk from L. monocytogenes in UCFM products was considered low, as it will not grow in the product, however it may survive for extended periods. The risk from Toxoplasma in UCFM products was also considered low, as the risk can be
considerably reduced if the meat is frozen before use, as this will kill the organism. Kebabs were predicted to be a significant source of illness if allowed to be recontaminated in the drip tray after cooking, with approximately 25,000 illnesses per annum. Under normal conditions for cooking kebabs, or even when an extra cook step was added, it was modelled that in 1% of cases, there may still be an element of undercooking or some other mishandling to allow survival of Salmonella. In this scenario, there were still 25 illnesses per annum predicted with kebabs as the cause. A survey of NSW kebab retail outlets (Jansson et al, 2008) showed that 92% of outlets undertook an extra cooking step after slicing the meat from the kebab. Table 20 – NZFSA risk profile outcomes examining hazards in processed meats Hazard

Listeria monocytogenes in

processed ready-to-eat meats
(Lake et al, 2002)
Shiga-like toxin producing
Escherichia coli in uncooked
comminuted fermented meat
products (Gilbert et al, 2007)

Food Safety Scheme Risk Assessment

Risk
Several notified cases in New Zealand and an outbreak of
non-invasive listeriosis in February/March 2000 associated
with corned silverside and ham indicate that processed RTE
meats are a route of infection for listeriosis in New Zealand. While uncooked comminuted fermented meat (UCFM)
products might appear to be a higher risk, well controlled
processing to lower the pH and water activity control
STECs.

Page 60 of 189

Table 21 – Risk ranking for processed meat products
Processed meat
product

Hazard

Severity

Probability

Growth
required
to cause
illness

Consumer
does
pathogen
reduction
step

Epidemiological
link

Risk
rating

Predicted
annual
number of
illnesses (in
Australia) 10;

Yes

Yes
Yes
Yes
Yes

Effect of
production,
processing,
handling on
the hazard
↓↑→
↓↑
↓↑
↓→
↓
↓

Deli meats
Terrines
Fresh sausage
Cooked sausages
Green runners
Kebabs

L. monocytogenes
L. monocytogenes
L. monocytogenes
L. monocytogenes
Salmonella

IB
IB
IB
IB
II/IB

Low
Low
Low
Low
N/A

No
No
Yes
No
Yes

Yes
Yes
No
No
No

High
Medium
Low
Low
Low

43
0.8
>0.01
0.04
0

Kebabs –
Salmonella
II/IB
Low

contaminated in
drip tray
Kebabs – normal
Salmonella
II/IB
Low
Kebabs – finished
Salmonella
II/IB
Low
with extra cook
step
Uncooked comminuted fermented meat (UCFM) products

Yes

↓→

Yes

Yes

High

27,500

Yes
Yes

↓→
↓→

Yes
Yes

Yes
Yes

Medium
Medium

28
28

Salami – general
population

Salami – vulnerable
population
Salami – unreliable
process

EHEC

IA/IB
II/IB
IB
IB

Low
Low
Low
Low

No
Yes
Yes
No

↓

↓
↓
↓freezing

No
No
No
No

Yes
Yes
No
No

Medium
Medium
Low
Low

1
12
>0.01
0

EHEC

IA/IB

Low

No

↓

No

Medium

125

EHEC

IA/IB

Low

No

↓

No

Yes – high level
contamination
Yes

High

604

Salmonella
L. monocytogenes
Toxoplasma
gondii

adapted from Sumner (2002); MLA (2003a)
10

Data from Sumner (2002) predicted annual numbers of illness per annum for the South Australia population (1.5 million), the MLA Risk Profile (MLA, 2003a)
used an Australian population figure of 19.7 million. These estimates have been extrapolated to the current population of Australia estimated by ABS (2009) as approximately 21.6 million, by multiplying by a factor of 14.3 and 1.1 respectively.

Food Safety Scheme Risk Assessment

Page 61 of 189

Table 22 – Risk ranking for L. m onocytogenes- contaminated processed meats Processed meat
product
Deli meats
Frankfurters, not
reheated
Pâté and meat spreads
Frankfurters reheated
Dry / semi-dry
fermented sausage

Risk
ranking
(per
serve)
High
High
High
Low
Low

Predicted cases of
listeriosis per
serve (in
Australia) 11
7.7 x 10-8

6.5 x 10-8

Risk
ranking
(per
annum)
Very high
High

Predicted
annual number
listeriosis cases
(in Australia) 12
133
2.5

3.2 x 10-8
6.3 x 10-11
1.7 x 10-11

Moderate
Low
Low

0.3
0.03
0

adapted from FDA/USDA (2003)

Conclusion
The production and processing of meat has a long history of successful regulation. The preventative programs implemented by government and industry have improved animal health to the point that many diseases are no longer present in Australian animals. The meat food safety scheme under the Food
Regulation 2004 requires compliance with national meat standards, leading to ante-mortem and post mortem inspections at abattoirs that have been effective in ensuring that meat produced in NSW is safe and suitable for human consumption. The microbiological surveys of meat production have shown a steady increase in the quality of meat produced. However, epidemiological data suggests that the prevalence of Salmonella and Campylobacter on raw poultry significantly contributes to the burden of foodborne illness within the community, not only from the consumption of contaminated poultry itself but the added potential for the introduction of these pathogens from poultry into food preparation areas where they may be a source of cross contamination onto RTE foods. Currently the food safety scheme requires compliance with the national poultry meat standard, however this only provides control measures for the processing sector. A whole chain approach is considered necessary, with control measures introduced at the primary production level to reduce the prevalence of these foodborne pathogens. With this aim in mind, FSANZ are currently finalising the development of Standard 4.2.2 – Primary Production and Processing Standard for poultry meat into Chapter 4 of the Food Standards Code. When finalised, this will be adopted into NSW legislation. The contamination of processed meats with L. monocytogenes continues to cause issues for meat processors, including a substantial number of product recalls. Additional hygiene and sanitation measures, including mandating the testing of finished product and food contact surfaces in high risk processing facilities, such as those packaging sliced RTE meats, aims to minimise contamination of this organism in product.

11

12

The risk per serving is inherent to the particular food category, and is therefore assumed to be the same in Australia as that calculated for the USA (FDA/USDA, 2003). This is based on the assumption that consumption patterns for these foods are identical in Australia and the USA. One significant difference would be that frankfurters are not commonly eaten without reheating in Australia, with MLA (2003b) estimating that only 5% are eaten
without further cooking

The risk per annum has been adapted from USA population data contained in the FDA/USDA (2003) risk assessment of 260 million and extrapolated to Australian population data of approximately 21.6 million (ABS, 2009) by dividing by a factor of 12

Food Safety Scheme Risk Assessment

Page 62 of 189

References – Meat and meat products
ABARE [Australian Bureau of Agricultural and Resource Economics] (2007). Australian

Commodity Statistics 2007. Canberra. Retrieved 27 November 2008, from

http://www.abareconomics.com/publications_html/acs/acs_07/acs_07.pdf.

ABS [Australian Bureau of Statistics] (1995). National Nutrition Survey: Foods Eaten, Australia, 1995. Australian Bureau of Statistics report. ABS Cat no 4804.0. Retrieved 13 January 2009, from
http://www.ausstats.abs.gov.au/ausstats/subscriber.nsf/0/CA25687100069892CA256888001C D460/$File/48040_1995.pdf
ACMF [Australian Chicken Meat Federation] (2003). National Biosecurity Manual Contract Meat Chicken Farming. Retrieved 14 January 2009, from

ACMF [Australian Chicken Meat Federation] (2008). Industry facts and figures. Retrieved 4 December 2008, from http://www.chicken.org.au/page.php?id=4. Adak, G.K., Meakins, S.M., Yip, H., Lopman, B.A. & O’Brien, S.J. (2005). Diseases risks from foods, England and Wales, 1996-2000. Emerging Infectious Diseases, 11, 365-72. Ampt, P. & Owen, K. (2008). Consumer attitude to kangaroo meat products. RIRDC Publication No 08/026. Retrieved 27 November 2008, from

http://www.rirdc.gov.au/reports/NAP/08-026.pdf.
Andrew, A.E. (1988). Kangaroo meat – Public health aspects. Australian Zoologist, 24(3), 138140. APL [Australian Pork Limited] (2008) Annual Report 2007-08. Retrieved 4 December 2008, from . Barton, M.D. & Robins-Browne, R.M. (2003). Yersinia enterocolitica. In Hocking A.D. (Ed.) Foodborne Microorganisms of Health Significance (pp. 577-596). Australian Institute of Food Science and Technology, Waterloo.

Bass, C., Crick, P. & Cusack, D. (2008). NSW domestic red meat abattoir evaluation – Final report. Retrieved 4 December 2008, from

http://www.foodauthority.nsw.gov.au/_Documents/science/nsw_domestic_red_meat_abattoir _evaluation.pdf.

Cassin, M.H., Lammerding, A.M., Todd, E.C.D., Ross W., McColl, R.S. (1998) Quantitative risk assessment for Escherichia coli O157:H7 in ground beef hamburgers. International Journal of Food Microbiology, 41(1), 21-44.

Cressey, P. Lake, R. & Hudson, A., & Nortje, G. (2006). Risk profile: Mycobacterium bovis in red meat. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/risk-profiles/FW0320_Mbovis_in_meat_final_May_2006.pdf DAFF [Department of Agriculture, Fisheries and Forestry, Australian Government of] (2004). Generic Import Risk Analysis (IRA) for pig meat – final import risk analysis report. Retrieved 5 December 2008, from Doolittle, R. (2008, December 18). Maple Leaf settles listeria suits for $27M. The Toronto Star Retrieved 19 January 2009, from http://www.thestar.com/business/article/555873 Eberhart-Phillips, J., Walker, N., Garrett, N., Bell, D., Sinclair, D., Rainger, W., & Bates, M. (1997). Campylobacteriosis in New Zealand: results of a case-control study.
Journal of Epidemiology and Community Health 51:686-691

EFSA [European Food Safety Authority] (2007). Geographical BSE Risk (GBR) assessment covering 2000-2006. List of countries and their GBR level of risk as assessed by the Scientific Steering Committee and the European Food Safety Authority (EFSA). Retrieved 14 January Food Safety Scheme Risk Assessment

Page 63 of 189

2009, from
www.efsa.europa.eu/cs/BlobServer/Scientific_Document/GBR_assessments_table_Overview_ assessed_countries_2002-2006,0.pdf
EFSA [European Food Safety Authority] (2008). A quantitative microbiological risk assessment on Salmonella in meat: Source attribution for human salmonellosis from meat. The EFSA Journal 625, 1-32.

FAO/WHO [Food and Agriculture Organization of the United Nations / World Health Organization] (2001). Risk characterization of Salmonella spp. in eggs and broiler chickens and Listeria monocytogenes in ready-to-eat foods. FAO Food and Nutrition Papers – 72. Joint FAO/WHO Expert Consultation on Risk Assessment of Microbiological Hazards in Foods. Retrieved 30 October 2008, from http://www.fao.org/docrep/008/y1332e/y1332e00.HTM FAO/WHO [Food and Agriculture Organization of the United Nations / World Health Organization] (2003). Risk assessment of Campylobacter spp. in broiler chickens and Vibrio spp. in seafood. FAO Food and Nutrition Papers – 75. Joint FAO/WHO Expert Consultation on Risk Assessment of Microbiological Hazards in Foods. Retrieved 30 October 2008, from FDA/USDA [Food and Drug Administration/ United States Department of Agriculture] (2003).

Quantitative assessment of relative risk to public health from foodborne Listeria monocytogenes among selected categories of ready-to-eat foods. Retrieved 30 October 2008, from .

Food Science Australia & Minter Ellison Consulting (2002). National Risk Validation Project. Final Report.
FSIS [Food Safety and Inspection Service, USDA] (2002). Comparative Risk Assessment For Intact And Non-Intact (Tenderized) Beef: Executive Summary. Retrieved 14 January 2009, from http://www.fsis.usda.gov/PDF/Beef_Risk_Assess_Report_Mar2002.pdf FRSC [Food Regulation Standing Committee] (2007a). Australian standard for construction of premises and hygienic production of poultry meat for human consumption. FRSC Technical Report No 1. CSIRO publishing www.publish.csiro.au

FRSC [Food Regulation Standing Committee] (2007b). Australian standard for the hygienic production of wild game meat for human consumption. FRSC Technical Report No 2. CSIRO

publishing www.publish.csiro.au

FRSC [Food Regulation Standing Committee] (2007c). Australia standard for the hygienic production and transportation of meat and meat products for human consumption. FRSC Technical Report No 3. CSIRO publishing www.publish.csiro.au FSANZ [Food Standards Australia New Zealand] (2005). Scientific Assessment of the Public Health and Safety of Poultry Meat in Australia is available. Retrieved 14 January 2009, from FSANZ [Food Standards Australia New Zealand] (2006). Public health and safety of poultry meat in Australia. – Explanatory summary of the scientific assessment., Canberra. Gilbert, S., Lake, R., Hudson, A. & Cressey, P. (2007). Risk profile: Shiga toxin-producing Escherichia coli in uncooked comminuted fermented meat products. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/riskprofiles/FW0611_STEC_in_UCFM_August_2007_Final.pdf ICMSF [International Commission on Microbiological Specifications for Foods] (2002). Microorganisms in Foods 7: Microbiological testing in food safety management. Kluwer Academic/Plenum Publishers, New York.

Jansson, E., Bird, P., Saputra, T. & Arnold, G. (2008) ‘Food Safety Survey of Retail Doner Kebabs in NSW’, Food Australia. 60 (3), 95-98.
Lake, R., Hudson, A. & Cressey, P. (2002a). Risk profile: Shiga toxin-producing Escherichia coli in red meat and meat products. Institute of Environmental Science and Research Limited Food Safety Scheme Risk Assessment

Page 64 of 189

report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/risk-profiles/stec-in-red-meat.pdf Lake, R., Hudson, A. & Cressey, P. (2002b). Risk profile: Toxoplasma gondii in red meat and meat products. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/risk-profiles/toxoplasma-gondii-in-red-meat.pdf

Lake, R., Hudson, A. & Cressey, P. (2004). Risk profile: Yersinia enterocolitica in pork. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from

http://www.nzfsa.govt.nz/science/risk-profiles/yersinia-in-pork.pdf Lake, R., Hudson, A., Cressey, P. & Gilbert, S. (2007a). Risk profile: Campylobacter jejuni/coli in poultry (whole and pieces). Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/risk-profiles/campylobacter.pdf Lake, R., Hudson, A., Cressey, P. & Gilbert, S. (2007b). Risk profile: Campylobacter jejuni/coli in Mammalian and Poultry Offals. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/riskprofiles/FW0465_Campy_in_Offal_PVDL_fina
l_comments_Mar_2007.pdf Lake, R., Hudson, A., Cressey, P. & Gilbert, S. (2007c). Risk profile: Campylobacter jejuni/coli in red meat. Institute of Environmental Science and Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/riskprofiles/FW0485_Campy_in_red_meat_Final_sent_to_NZFSA_Jan_07.pdf Lake, R., Hudson, A., Cressey, P. & Nortje, G. (2002). Risk profile: Listeria monocytogenes in processed ready-to-eat meats. Institute of Environmental Science and Research Limited

report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/risk-profiles/listeria-in-rte-meat.pdf Lake, R., Hudson, A., Cressey, P. & Wong, T. & Gilbert, S. (2004). Risk profile: Salmonella (non typhoidal) in poultry (whole and pieces). Institute of Environmental Science and

Research Limited report prepared for the New Zealand Food Safety Authority. Retrieved 14 January 2009, from http://www.nzfsa.govt.nz/science/data-sheets/salmonella-poultryupdate.pdf

Leech, A., Shannon, P., Kent, P., Runge, G., Warfield, B. (2003) Opportunities for Exporting Game Birds. Rural Industries Research and Development Corporation (RIRDC). Report

Number 03/106. Retrieved 6 November 2009, from http://www.rirdc.gov.au/reports/NAP/03106.pdf McEvoy, J.M., Sheridan, J.J. & McDowell, D.A. (2004). Major pathogens associated with the processing of beef. In Smulders F.J.M & Collins, J. D. [Eds] Food safety assurance and veterinary public health Volume 2. Safety Assurance during food processing pp 57-80. Wageningen Academic Press. , Wageningen Netherlands

Meat Standards Committee (2002). Microbiological testing for process control in the meat industry – guidelines. Retrieved 19 February 2009, from

Meat Standards Committee (2008). Regulatory guidelines for the control of Listeria. In NSW Food Authority (2008) Listeria management program.

Food Safety Scheme Risk Assessment

Page 65 of 189

MLA [Meat & Livestock Australia] (2000). The Microbiology of Australian Meat (1998). North Sydney: Meat and Livestock Australia.
MLA [Meat & Livestock Australia] (2003a). Through chain risk profile for the Australian red meat industry, PRMS.038c, part 1: risk profile. North Sydney: Meat and Livestock Australia. MLA [Meat & Livestock Australia] (2003b). Guidelines for the safe manufacture of smallgoods. North Sydney: Meat and Livestock Australia.

MLA [Meat & Livestock Australia] (2005). Microbiological quality of Australian beef and sheepmeat – results of the industry’s third national abattoir study. North Sydney: Meat and Livestock Australia.

MLA [Meat & Livestock Australia] (2006). Listeria monocytogenes in smallgoods: Risks and controls. North Sydney: Meat and Livestock Australia.
MLA [Meat & Livestock Australia] (2008a). Fast facts 2008. Australia’s beef industry. Retrieved 14 January 2009 from, http://www.mla.com.au/NR/rdonlyres/3EF73ECB-4FBB4455-A561-14CC636D7ADB/0/BeefFastFacts2008.pdf MLA [Meat & Livestock Australia] (2008b). Fast facts 2008. Australia’s sheepmeat industry. Retrieved 14 January 2009, from http://www.mla.com.au/NR/rdonlyres/70C05D4D-5D3E425D-96F7-4B973E5BFF55/0/SheepmeatFastFacts2008.pdf NSW Food Authority (2008). Listeria management program. Retrieved 19 February 2009, from http://www.foodauthority.nsw.gov.au/_Documents/industry_pdf/listeria-managementprogram.pdf PISC [Primary Industries Standing Committee] (2006). Model Code of Practice for the Welfare of Animals – Land Transport of Poultry. PISC Report 91.

Pointon, A., Sexton, M., Dowsett, P., Saputra, T., Kiermeier, A., Lorimer,
M., Holds, G., Arnold, G., Davos, D., Combs, B., Fabiansson, S., Raven, G., McKenzie, H., Chapman, A. & Sumner, J. (2008). A baseline survey of the microbiological quality of chicken portions and carcasses at retail in two Australian states (2005 to 2006). Journal of Food Protection, 71, 1123-1134.

Pople, T. & Grigg, G. (1999). Commercial harvesting of kangaroos in Australia – background paper. Retrieved 10 November 2008, from

Ross, T. & Shadbolt, C.T. (2001). Predicting pathogen inactivation in uncooked comminuted fermented meat products. Report for Meat & Livestock Australia. Sumner, J. (2002). Food Safety Risk Profile for Primary Industries in South Australia (Final Report). Department of Primary Resources SA, Adelaide. Retrieved 30 September 2008, from

Wallace. R.B. (2003). Campylobacter in. In Hocking A.D. (Ed.) Foodborne Microorganisms of Public Health Significance (pp. 311-331). Australian Institute of Food Science and

Technology, Waterloo.

Widders, P.R., Coates, K.J., Warner, S., Beattie, J.C., Morgan, I.R. & Hickey, M.W. (1995). Controlling microbial contamination on beef and lamb during processing. Australian Veterinary Journal, 72(6), 208-211.

WHO/FAO [World Health Organization/Food and Agriculture Organization of the United Nations] (2002). Risk assessment of Salmonella in eggs and broiler chickens: Interpretive summary. Retrieved 14 January 2009, from

www.who.int/foodsafety/publications/micro/salmonella/en/index/html

Food Safety Scheme Risk Assessment

Page 66 of 189

Plant products food safety scheme
Hazard identification
In 2000, the former SafeFood Production NSW commissioned Food Science Australia to determine the relative food safety risks for various plant products produced and/or marketed in NSW (FSA, 2000a). This work resulted in six products being ranked as high risk due to microbiological hazards (Table 23), and formed the scientific basis for the introduction of the Plant products food safety scheme into the Food Regulation 2004. The scheme was developed to introduce minimum regulatory requirements for businesses producing high risk plant products, and to implement control measures to minimise the risks from the microbiological hazards associated with these products.

Table 23 – Microbiological hazards associated with plant products Plant product
Fresh cut vegetables – may be
consumed raw
Fresh cut vegetables – chilled,
MAP or extended shelf life
Vegetables in oil
Seed sprouts
Fresh cut fruit
Fruit juice / drink (unpasteurised)

High risk ranking
Pathogenic E. coli
Salmonella serovars

L. monocytogenes
L. monocytogenes
C. botulinum
C. botulinum

Pathogenic E. coli
Salmonella serovars
Pathogenic E. coli
Salmonella serovars
L. monocytogenes
Salmonella serovars
Pathogenic E. coli

Medium risk ranking

B. cereus
L. monocytogenes
Cryptosporidium parvum
Enteric viruses

adapted from FSA (2000a)

Fresh cut vegetables
Fresh cut fruits and vegetables are raw agricultural products that have been processed by means of washing, trimming, cutting or slicing to make them ready for consumption. Contamination of vegetables may occur during growth, harvest, or processing. Under certain conditions microorganisms can also become internalised within the vegetables. Conditions that promote internalisation of microorganisms include damage to the natural structure (eg punctures, stem scars, cuts, splits) and placing warm produce into cooler, contaminated wash water.

The actual process of cutting and/or removing the protective outer surfaces of the plants may increase the potential for pathogenic bacteria to survive and/or grow. Many vegetable products do not undergo a kill step that will completely eliminate pathogens, however measures such as sanitising washes may be used to reduce microbial contamination of pathogens.

Many fresh cut vegetables are packaged under modified atmosphere packaging (MAP) and refrigerated to extend the shelf life. This form of processing may
lead to an increased risk with pathogens such as L. monocytogenes and psychrotrophic strains of C. botulinum by enhancing the conditions for their survival and allowing additional time for growth. MAP products may become fully anaerobic if the plant tissue is actively respiring and uses up all the oxygen. As any competition from aerobic spoilage organisms is inhibited, this may increase the opportunity for anaerobic or facultative anaerobic pathogens to grow.

Food Safety Scheme Risk Assessment

Page 67 of 189

Fresh cut fruit
Fresh fruit are normally perceived as low risk foods, as they tend to have a thicker protective skin than most vegetables and most are harvested from trees or bushes. The notable exceptions are melons and strawberries, which are considered higher risk because they grow close to the ground and their surfaces may become contaminated with soil.

Contamination of fruit may occur at any point from growing (soil, fertilisation, irrigation water, animal/bird waste), through to harvesting, processing (including washing), distribution, marketing and consumption. Many microbial pathogens cannot survive or grow on most fruit due to the low pH environment. However, melons and strawberries have relatively high pH which makes them more likely to be a food safety hazard. In addition, the skin of rockmelon tends to be porous which may allow the penetration of pathogens and agricultural chemicals into the fruit. Melons are often dipped in a sanitising solution after harvest (FSA, 2000a). Fresh cut fruits may be value added by peeling, chopping, slicing and packaging (FSA, 2000a). Many fresh cut fruit are packaged under MAP and refrigerated to extend the shelf life. With additional time, this can lead to an increased risk from pathogens that are adapted to the acidic environment of fruit and are able to survive and grow in these foods.

A range of bacterial and viral pathogens and enteric parasites have been
identified as being of concern in fresh cut fruit. The actual process of cutting and/or removing protective outer surfaces of the fruit may increase the potential for pathogens to survive and/or grow. Fruit pickers and handlers with infections are also an important source of contamination.

Vegetables in oil
This product category includes a diverse range of vegetables and mixtures of vegetables and herbs that may be used fresh, dried, roasted or acidified. Oil is added to exclude air, which prevents discolouration of the vegetable. Although immersion of vegetables in oil reduces the available oxygen in the container, contrary to popular belief, it does not preserve the food. Some pathogenic bacteria are able to survive and grow in reduced levels of oxygen and even under anaerobic conditions in the absence of oxygen.

C. botulinum is the main pathogen of concern because of its ability to grow anaerobically and it has been linked to outbreaks of illness from the consumption of vegetables in oil. Vegetables may be contaminated by C. botulinum spores, which are frequently associated with soil and processes such as cooking and acidification may be insufficient to inactivate the spores or prevent their germination and growth. Acidification to below pH 4.6 should prevent outgrowth, however more than one hurdle is recommended as a safeguard.

Seed sprouts
Seed sprouts are usually consumed raw and include alfalfa, mung bean, chickpeas, cress, fenugreek, soy, lentils, sunflower, onion and radish. Seeds for sprouting generally do not receive any special treatment during harvesting and transport and so may become contaminated with pathogenic organisms in the field or during harvesting, handling, processing and distribution. While some bean sprouts may be cooked prior to consumption, many others are consumed raw, for instance with salads.

Food Safety Scheme Risk Assessment

Page 68 of 189

The microbiological pathogens frequently found associated with seeds for sprouting include B. cereus, Salmonella serovars, and E. coli and these organisms have also been implicated in foodborne illness outbreaks. The rough surfaces and cracks in the seed may protect the pathogens from microbiocidal treatments and may make detection during routine analysis difficult. High levels of organic matter also reduce the effectiveness of chlorine treatments during seed washing and seed sprouting. Bacterial populations of 102-107 cfu/g have been observed on seeds for sprouting and this natural population can rapidly increase under the high moisture and moderate temperature conditions used in sprouting facilities. Microorganisms may also become internalised in the sprout during growth so sanitising wash treatments of sprouted seeds are not likely to be effective (FSA, 2000a). Unpasteurised fruit juice

Fruit juices are made by extracting fruit (citrus juices) or by macerating fruit (grape, cherry, berry, apple juice, etc). This may be followed by clarification, filtration, pasteurisation, and/or other processes to reduce the microbial load. In recent years there has been a trend to produce ‘natural’ fruit juices containing no preservatives and receiving little or no heat treatment.

Any microorganisms present on the surface of fruit may potentially contaminate the juice made from it. Bacterial pathogens are unlikely to grow due to the low pH but some bacteria, viruses or protozoa may be able to survive for extended periods. The length of time the microorganism may survive is dependent on the pH of the juice, storage temperature and the physiological state of the microorganism. Some Salmonella serovars and strains of pathogenic E. coli are known to be particularly acid tolerant, with this response thought to be activated by previous exposure to sub-lethal pH values.

Apple and pear juice are can become contaminated by the mycotoxin patulin which is produced by several Penicillium and Aspergillus species. P. expansum appears to be the main patulin producer in apples and apple
products. Since patulin is concentrated in the rotting tissue of fruit, it is a good indicator of the quality of fruit used to make the juice.

The acidic nature of fruit juices makes them corrosive to metals. To avoid potential chemical contamination, only stainless steel or corrosion resistant vessels should be used to store these products. Other metals such as copper can leach into the beverage during storage.

Exposure assessment
Production data
Leafy salad vegetables, such as lettuce, rocket and baby spinach are the most common products in the fresh cut category, contributing towards an estimated national production value of $44 million for the year 1997–98 (Szabo & Coventry, 2001).

The Regulatory Impact Statement (RIS) prepared for the Food Regulation 2004, based on limited industry information, estimated annual NSW consumption of the fresh cut fruit and vegetables (NSW Food Authority, 2004) as: •

11,000 tonnes fresh cut vegetables, with a high proportion imported from Victoria and Queensland

•

150 tonnes of fresh cut fruit

Food Safety Scheme Risk Assessment

Page 69 of 189

•

Approximately 1000 tonnes of vegetables in oil, with the vast majority imported from overseas and interstate and

•

Between 2100 and 2600 tonnes of seed sprouts

Subsequent recent surveys by the NSW Food Authority of the NSW sprout industry suggest that 2007/08 production was in the order of 3630 tonnes, with some of this product being sold interstate. NSW fruit juice suppliers suggest that manufacture of unpasteurised fruit juices occurs at relatively low volume, about 100,000 L/year, not including juices prepared in retail premises (NSW Food Authority, unpublished). Consumption of plant products

Consumption data for fruits and vegetables from the National Nutrition Survey are summarised in Table 24 (ABS, 1995). During the period 1997-98 and 1998-99, fruit and fruit products (including fruit juices) consumption increased by 8.3% from 124.7 kg per capita to 135.0 kg. In the same period, imports for oranges and other citrus fruit rose by more than 62% (ABS, 2000).

Consumption of vegetables has shown a steady 9.4% increase over the last decade. Per capita consumption of tomatoes showed a significant increase from 20.9 kg in 1997-98 to 24.9 kg in 1998-99, a rise of 19%. The category of other vegetables showed an increase in consumption in 1998-99 of 4.6% to 25.1 kg per person. Data from the Australian 1995 National Nutrition Survey (ABS, 1995) indicates that fruit juices and drinks are consumed in significant amounts by a large proportion of the population. Approximately 35% of all respondents consumed fruit juices and drinks with the mean consumption being 250mL per day.

Food Safety Scheme Risk Assessment

Page 70 of 189

Table 24 – Consumption of fruits and vegetables in Australia Sex

Age

Male
Male
Male
Male

2-3
4-7
8 – 11
12 15
16 18
19 24
25 44
45 64
65+
2-3
4-7
8 – 11
12 15
16 –
18
19 24
25 44
45 64
65+

Male
Male
Male
Male
Male
Female
Female
Female
Female
Female

Female
Female
Female
Female

Proportion
of persons
consuming
fruit
products and
dishes 13
(%)
77.6
65.6
56.4
49.9

Median daily
intake for
consumers of
fruit products
and dishes
(g/day)
153.6
168.0
166.0
167.2

Proportion of
persons
consuming
vegetable
products and
dishes 14
(%)

68.1
72.7
77.0
78.8

Median daily
intake for
consumers of
vegetable
products and
dishes (g/day)
92.1
118.0
165.0
223.0

39.9

172.0

83.1

253.6

31.9

179.2

84.7

271.3

45.8

210.0

86.6

263.0

59.5

229.0

91.0

297.8

69.6
75.4
72.8
62.5
58.0

202.0
140.0
166.0
150.8
172.0

91.7
79.2
79.7
77.0
85.9

280.4
95.2
122.5
158.0

180.8

41.1

191.0

85.8

185.0

41.4

166.0

86.5

220.5

55.0

188.4

88.0

216.1

69.8

192.0

91.0

258.5

75.6

196.0

91.5

239.3

adapted from National Nutrition Survey (ABS, 1995)

The consumption data does not provide information on how much unpasteurised juice is consumed, however the Australian Fruit Juice Association believes that approximately 95% of juice sold has undergone some form of pasteurisation process.

13

Fruit products and dishes are defined in the National Nutrition Survey (ABS, 1995) as including the following: – Pome fruit, berry fruit, citrus fruit, stone fruit, tropical fruit, other fruit – Mixtures of two or more groups of fruit

– Dried fruit, preserved fruit
– Mixed dishes where fruit is the major component
14
Vegetable products and dishes are defined in the National Nutrition Survey (ABS, 1995) as including the following:
– Potatoes, cabbage, cauliflower and similar brassica vegetables, carrot and similar root vegetables, leaf and stalk vegetables, peas and beans, tomato and tomato product, other fruiting vegetables – Other vegetable and vegetable combinations

– Dishes where vegetable is the major component

Food Safety Scheme Risk Assessment

Page 71 of 189

Prevalence of hazards in plant products
There have been a small number of surveys of Australian plant products. Arnold & Coble (1995) in a broad survey of NSW food found 1/54 samples (1.9%) of RTE salads and vegetables positive for L. monocytogenes.

Szabo et al (2000) tested 120 minimally processed, cut and packaged lettuce samples. Three samples (2.5%) were positive for L. monocytogenes, 66 samples (55%) were positive for Aeromonas hydrophila or A. caviae and 71 samples (59%) were positive for Y. enterocolitica.

The Victorian Department of Human Services (DHS) surveyed the microbiological quality of freshly squeezed juices from retail businesses across the state (Victorian DHS, 2005). L. monocytogenes was detected in 1/291 samples (0.3%), but the level was sufficient to classify the sample as potentially hazardous. E. coli was detected in 7/291 samples (2.4%). Salmonella and coagulase positive S. aureus were not detected.

The Western Australian Department of Health (WA Health, 2006) tested 261 samples of sprouted seeds from retail stores. E. coli was detected in 7 samples (2.7%), while Listeria and Salmonella were not detected in any samples.

The NSW Food Authority undertook a small survey in 2006 to determine the safety of fresh cut vegetables sold in NSW. E. coli was detected in 1/119 samples (0.8%), while Salmonella, L. monocytogenes and verotoxigenic E. coli (VTEC) were not detected in any samples (NSW Food Authority, unpublished).

The Authority has also undertaken several surveys of seed sprouts. In 2005, all 30 samples were found to be microbiologically acceptable. In 2006, 1/36 samples (2.7%) was found to be potentially hazardous due to the presence of VTEC and a further two samples were categorised as unsatisfactory due to elevated levels of E. coli. A more extensive survey in 2008 of 122 samples found 99.2% of samples were microbiologically acceptable, with a single
sample categorised as unsatisfactory due to B. cereus at a level of 5500 cfu/g (NSW Food Authority, 2008). There has been considerable international interest in the safety of plant products. O’Brien et al (2000) prepared a discussion paper on the microbiological status of RTE fruit and vegetables for the UK Advisory Committee on the Microbiological Safety of Food (ACMSF). The report provided a summary of foodborne illness outbreaks and surveys of plant products. The report concluded that while contamination of raw vegetables usually occurs at low prevalence, it is pervasive. The Food Safety Authority of Ireland (FSAI) surveyed the bacteriological safety of a range of plant products as part of a European Commission coordinated program (FSAI, 2003). Pre-cut fruit and vegetables had samples classed as unacceptable/potentially hazardous due to the presence of Salmonella in 1/529 samples (0.2%) and L. monocytogenes in 1/344 samples (0.3%). Qualitative tests found 21/513 samples (4.1%) positive for L. monocytogenes. No sprouted seeds samples were classed as unacceptable or potentially hazardous. L. monocytogenes was detected in 1/26 samples (3.8%). No problems were detected with unpasteurised fruit and vegetable juices.

A similar European Commission program surveyed pre-packed mixed salads from retail premises in the UK for L. monocytogenes (Little et al, undated). L. monocytogenes was detected in 4.8% of samples collected. A parallel survey by the FSAI included Salmonella testing in the survey design (FSAI, undated). Qualitative analysis showed L. monocytogenes was present in 19/714 samples Food Safety Scheme Risk Assessment

Page 72 of 189

(2.7%). Quantitative analysis detected two samples with L. monocytogenes at levels exceeding 100 cfu/g, however Salmonella was not detected in any sample. Little & Gillespie (2008) summarised microbiological results of surveys of prepared salads and fruit examined in the UK. No isolations of E. coli O157 or Campylobacter were reported. Five of 3852 samples (0.1%) of bagged salad vegetables were positive for Salmonella but other commodities were negative. L. monocytogenes and E. coli were detected in most commodities surveyed, usually at low incidence. Crepet et al (2007) used
statistical techniques on 165 prevalence studies and concentration data from 15 studies of L. monocytogenes in fresh vegetables to estimate an overall probability of significant counts being found in the products. Their mathematical method required a minimum of one positive sample in each survey. However, as some survey sets had no actual detections, this change was made to data sets to accommodate the statistical analysis. Acknowledging this deliberate overestimation, the authors calculated the probability of sample contamination with L. monocytogenes exceeding 10 cfu/g as 1.4%, exceeding 100 cfu/g as 0.6% and exceeding 1000 cfu/g as 0.2%.

Hazard characterisation
Foodborne illness outbreaks from plant products
An indication of the exposure to hazards in plant products is provided by an examination of the Australian foodborne illness outbreaks between 1995 and 2008 attributed to fresh produce and plant products, summarised in Table 25 (details of each outbreak are included in Table 67 of Appendix 3). Prior to this, two other Australian outbreaks of significance brought plant products into the spotlight as a significant source of foodborne illness. In NSW in 1989 there were three separate outbreaks from fruit salad due to Salmonella Bovismorbificans traced to a single NSW salad manufacturer (Biffin & McCarthy, pers comm), while in 1991 a nationwide outbreak from Norovirus was attributed to the consumption of unpasteurised orange juice. The juice was served on airline flights and was responsible for more than 3000 cases of illness (Foodlink, 2002). In addition, the risk of listeriosis from plant products was highlighted by an outbreak of listeriosis from contaminated fruit salad in NSW aged care facilities and hospitals in the Hunter Valley area. Through 19981999, six deaths of elderly patients occurred and nine were affected (this outbreak is included in outbreak data for the section on the Vulnerable persons food safety scheme).

Internationally, there have been many examples of outbreaks that have been attributed to plant products. Sivapalasingham et al (2004) summarised the outbreaks attributed to fresh produce in the USA from 1993 to 1997. The authors identified 190 produce-associated outbreaks, resulting in 16,058 illnesses, 598 hospitalisations and eight deaths. They report that
produce-associated outbreaks were an increasing proportion of all reported foodborne outbreaks with a known food cause, rising from 0.7% in the 1970s to 6% in the 1990s. Salad, lettuce, juice, melon, sprouts, and berries were the fresh produce most frequently implicated. Sivapalasingham et al (2004) also recognised Cyclospora and E. coli O157:H7 as novel causes of foodborne illness from plant products.