Food Control 22 (2011) 1674e1688

Contents lists available at ScienceDirect

Food Control
journal homepage: www.elsevier.com/locate/foodcont

HACCP-based program for on-farm food safety for pig production in Australia
Peter M. Horchner a, Andrew M. Pointon b, *
a b

Symbio Alliance, PO Box 4312, Eight Mile Plains, QLD 4113, Australia South Australian Research and Development Institute, GPO Box 397, Adelaide, SA 5001, Australia

a r t i c l e i n f o
Article history: Received 10 November 2010 Received in revised form 21 March 2011 Accepted 26 March 2011 Keywords: HACCP Food safety On-farm Pig production

a b s t r a c t
The standard Codex HACCP approach was modied to allow a hazard analysis and critical control point determination to be conducted at an industry level and then used to determine the appropriate on-farm food safety control measures for pig production in Australia. A detailed risk-based prole with hazard identication, hazard characterisation and levels of microbial contamination for production and primary processing was used as a major technical resource to inform HACCP determinations. The process resulted in the identication of Critical Control Points for control of a specic physical hazard (non-recovered broken needles) and prevention of violations of Maximum Residue Limits with agricultural and veterinary chemicals. In relation to the identied microbiological hazards, it was noted that there are numerous Good Agricultural Practices (GAP) to prevent and/or reduce their risk and these controls would also need to be a feature of any on-farm food safety program. By applying a HACCP-based approach it was determined that the application of a set of Good Agricultural Practices on-farm would be effective in ensuring low risk. It was, therefore, concluded that on-farm food safety programs may not warrant full (i.e. Codex compliant) HACCP plans at the individual enterprise level provided appropriate GAP is in place. The results provide pig producers and the Australian pig industry with the elements of a HACCPbased food safety system that are scientically justiable, understandable and realistic to apply. These features are essential elements that underpin successful implementation and compliance by industry. Crown Copyright 2011 Published by Elsevier Ltd. All rights reserved.

1. Introduction Until the early 1970s the size of the Australian national pig herd had remained fairly stable with most farms being farrow-to-nish i.e. the entire pork production cycle. However, since then the number of producers has dropped from approximately 39,000e1900 in 2007, with nearly an additional 500 producers growing out pigs that have been sourced from other pig production enterprises (Australian Pork Ltd, 2008). In 2007, there were 5.3 million pigs slaughtered for human consumption with pig meat production at approximately 377,000 t (Australian Pork Ltd, 2008). The estimated Gross Value of Production (GVP) for Australian pork production was $944 million for the period 2006e2007 (Australian Bureau of Statistics, 2008) decreasing to $895 million for the period 2007e2008 (Australian Bureau of Statistics, 2010). The Australian pork industry services the entire domestic fresh pork market and approximately 40% of the domestic processed market. Australia is both an exporter and importer of pork, with Singapore, Japan and New Zealand being the main export markets.

* Corresponding author. Tel.: 61 8 83039434; fax: 61 8 83039669. E-mail address: andrew.pointon@sa.gov.au (A.M. Pointon).

Canada, Denmark and the United States account for 99% of imported pork. With exports decreasing to 45,500 t (value $142.6 m) and imports increasing to 106,900 t (value $445 m) in 2008 (Australian Pork Ltd, 2008), Australian pork producers must ensure they remain competitive in the global market. Generic frameworks for managing foodborne risks to human health have been proposed by the Codex Alimentarius Commission (2002). These principles are reected in the Code of Hygienic Practice for Meat (Codex Alimentarius Commission, 2005). They require consideration of risk management of hazards prior to slaughter and during processing, based on the application of HACCP (Codex Alimentarius Commission, 2003b). Subsequently, in 2003, the Australian red meat industry undertook a risk prole (Codex Alimentarius Commission, 2003a) and a modied Codex HACCP approach to determine appropriate on-farm food safety controls and an underpinning of the industrys on-farm food safety programs (Horchner, Brett, Gormley, Jenson, & Pointon, 2006). Consistent with this approach, a risk-based prole of Australian pork products was developed to provide the background information (Pointon & Horchner, 2010) required to develop this on-farm HACCP plan. In this context the risk-based prole report provided technical rigour for the hazard analysis (Principle 1) of a HACCPbased system for the pork primary production sector.

0956-7135/$ e see front matter Crown Copyright 2011 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2011.03.028

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

1675

The purpose of this paper is to describe the application of HACCP at an industry level to determine appropriate on-farm food safety control measures applicable at the enterprise level for pig production. While the Australian Pig Industry Quality (AQIS, 2009) and PigPass quality assurance programs ofcially meet the requirements for on-farm control programs for chemical residues there is no formal package of controls for microbial hazards which has independent review or audit. Furthermore, in the current development of national Primary Production and Processing Standards for meat by Food Standards Australia New Zealand (FSANZ), Salmonella spp., Yersinia enterocolitica, Campylobacter jejuni/coli and Toxoplasma gondii have been identied as hazards associated with pork (FSANZ pers. comm.). This paper provides the technical basis for the national pork industrys primary production food safety program. 2. Methodology 2.1. HACCP approach The term HACCP plan implies that Codex HACCP methodology should be used (Codex Alimentarius Commission, 2003b) with the rst step being the formation of a HACCP team. The HACCP team modied the conventional approach to preparation of a HACCP plan (Codex Alimentarius Commission, 2003b) in order to meet the needs of this project. Specically, since HACCP had to apply to a large number of pork producing enterprises, its application was at an overall livestock sector level rather than an individual enterprise level. In addition, consideration was given to control measures and interventions further along the food chain that deal with hazards identied as being introduced on-farm, where applicable (Pointon & Horchner, 2010). The HACCP plan had to be broad enough to cover all enterprises and therefore, specic aspects may not necessarily pertain to a given enterprise. The team took a rigorous approach to the application of HACCP following the Logic Sequence for the Application of HACCP (Codex Alimentarius Commission, 2003b). The study addressed recognised and potential food safety hazards that can cause disease as a result of eating pork or pork products. Foodborne hazards considered included biological hazards, including attributes related to wholesomeness (microorganisms, gross carcase abnormalities), and physical (foreign objects and materials) and chemical agents (natural toxins, residues, heavy metals) with the potential to cause adverse health effects. Biological hazards included microbiological (e.g. Salmonella) and gross carcase lesions (i.e. abnormalities resulting from organisms or pathology associated with certain animal parasites and disease). Chemical hazards included heavy metals, biological toxins and residues from farm chemicals in the environment and/or in processing including those with an established Maximum Residue Limit (MRL) and/or Export Slaughter Interval (ESI) in place. Physical hazards considered were those which may enter during the production phase, especially broken needles. Potential hazards included those that may result in public health, social and/or economic impact but for which epidemiological evidence is lacking e.g. chemicals and toxins (OECD/WHO, 2003). Food safety threats to market access may or may not be valid hazards, but are potential technical barriers to trade (e.g. certain chemical residues that have no known adverse health effect). Stock feeds possibly containing genetically modied (GM) grains were not an issue at the time of conducting this study but were recognised by the HACCP team as an emerging issue which may need to be reconsidered in the future. 2.2. HACCP team (Step 1) In order to develop the HACCP plan an experienced HACCP team of seven members with specialist training in HACCP methodology,

veterinary public health, microbiology, epidemiology of foodborne diseases, pig production and pork processing was assembled. Members of the team were selected on the basis of having experience in the development of HACCP systems for a range of food and agricultural commodities, implementation of hazard prevention procedures, the application of risk assessment, associated food safety research, auditing food safety systems and livestock production including pigs. 2.3. Scope, product description and intended use (Steps 2 and 3) The scope of this HACCP study was limited to production of pigs for Australian and international markets with the intended use of meat consumption by the general population. It covered the process from inputs (animals, feed, water, treatments, site, husbandry practices) through to dispatch from property (Fig. 1). It also covered diverse production systems with respect to bedding and housing. Transport is covered by sector specic programs and was not within the scope of this study. The hazards of interest were identied as well as potential foodborne hazards and threats to market access linked to the production of pigs intended for human consumption (Pointon & Horchner, 2010). The scope of the hazard analysis also included steps beyond the on-farm livestock production process including downstream processing operations as recommended by Codex (Codex Alimentarius Commission, 2003b). Account was taken of the existence, or otherwise, of effective downstream control/reduction measures for hazards identied onfarm, but did not include interventions available to consumers (e.g. cooking). The HACCP plan was intended to provide technical underpinning for on-farm food safety programs for the Australian pig industry.
Selection of Breeding Stock and Mating

Culled Stock Sale of Breeders

Breeding

TRANSPORT Inputs - feeds - ag &vet chemicals - water - treatments - semen - purchase breeders - bedding Farrowing Outputs - effluent - dead pigs - sale of breeders Weaning

Grow/Finish

Weaner Grow/Finish Farm

TRANSPORT

Saleyard

Abattoir

Fig. 1. Overview of pig production processes.

1676

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

2.4. Process owchart (Steps 4 and 5) An overview of the livestock production process is provided in Fig. 1. Further details on the activities associated with each of these general steps, including any husbandry differences between enterprises, are provided in Table 1. Both Fig. 1 and Table 1 were developed to provide a generic framework to cover all enterprise types and production systems within the scope. Verication of this process owchart (Step 5) and the activities in Table 1 were undertaken by desktop review by industry stakeholders, including industry representatives of large and small growers, a major pork grower and processor and members of the national industry body (Australian Pork Ltd). 2.5. Principle 1 e conduct a hazard analysis: identication of hazards and potential hazards (Step 6) The hazard analysis, Principle 1 (Codex Alimentarius Commission, 2003b) was conducted at two levels e for pork
Table 1 Activities associated with each pig production (process) step. Process step Selection of breeding stock Associated activity

products and the pork industry overall and within an individual pig production enterprise to later determine Critical Control Points (CCPs). The hazard analysis was more extensive than that which an individual pig producing enterprise would undertake in order to establish a technical reference for individual producers and the pig production sector. The extensive technical review included classication of hazards, foodborne or otherwise hazard identication and characterisation in terms of primary source and human health consequences (Pointon & Horchner, 2010); hazard identication in relation to sectors of the pork industry (Fig. 2); and nally a hazard analysis within a pig production enterprise, including risk assessment and control measures required. 2.5.1. Identication and classication of hazards in pork and pork products Identied and potential microbial foodborne hazards are listed in the companion risk-based prole of Australian pork (Pointon & Horchner, 2010). These are drawn from a review of the literature and the published public health surveillance data in Australia (OzFoodNet, www.ozfoodnet.gov.au).

Husbandry differences Natural/induced oestrus

Breeding

Farrowing

Weaners

Growers/Finishers

Inputs

Select breeders e identication, traceability, vaccination, oestrus stimulation Purchase breeders e biosecurity, identication, traceability Acclimatisation e vaccination, environ exp., animal treatments Transfer breeders Mating Gestation Cull breeders, vaccination, health treatments Housing Production system Teeth/tails/identication Wean/foster Vaccination Health treatments Housing Castration Transfer stock Production system Vaccination Health treatments Housing Transfer Housing Health treatments Market selection Transport preparation Transport Feed Grain Protein meals Supplements Water Ag/vet chemicals

Natural, AI Euthanasia, sale Pens, stalls, outdoor, shelters, bedding systems Continuous/batch Ear notch, tattoo, RFID

Crates/outdoor

Continuous/batch

Off-site, pens, shed, shelters, bedding systems, outdoor Off-site, shed, shelters, bedding systems, outdoor

Tattoo, brand Home mixed, pelleted, by-products

Mains, bore, dam, river Sanitisers, pest control, animal health, herbicides, processing, grain treatments, Straw, rice hulls, sawdust Prior land use e outdoor, pest exposure, chemical drift, foreign bodies Personal hygiene Mulch from shelters

Treatments Bedding Semen Uncontrolled exposure Pre-requisite programs Personnel Efuent disposal Dead pig disposal Sale of breeders Selection of pigs for sale

Outputs

Sale pigs

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

1677

Q1. Is the hazard a food safety related market access requirement?

YES/NO Q2. Does the hazard cause foodborne illness?

YES Q3. Are animals a recognised source of the hazard in meat? NO

YES Q4. Are effective* measures available on farm to prevent, eliminate or reduce the hazard to an acceptable level?
(*effective means scientifically valid, practical and have positive benefit-cost ratio vs controls that can effectively be applied elsewhere)

NO

DO NOT INCLUDE in On Farm Food Safety Scheme. Consult vet or public health officials YES

Table 1. This second phase of hazard analysis was conducted for each of the hazards which remained relevant to the production sector after the assessment completed and summarised in Table 2 and Fig. 2. The process identied any on-farm control measures that could prevent, eliminate or reduce the hazard to an acceptable level (Pointon & Horchner, 2010) and the steps or activities within a livestock enterprise at which controls could be applied. Predisposing factors which inuence the occurrence of the potential hazard were identied based on the HACCP team members knowledge of livestock operations and hazards and the results of the previous industry level hazard analysis (Pointon & Horchner, 2010). A suitable method of risk assessment was developed (Table 4) using a combination of commonly used tools (Anonymous, 2005; ICMSF, 2002). 2.6. Principle 2 e determine the critical control points (Step 7)

Q5. Is the hazard controlled elsewhere? e.g. Food safety NO

schemes in other sectors or customer requirements

YES INCLUDE# in On Farm Food Safety Scheme

(#Use HACCP-based approach to determine HOW)

NO Further Information Required

e.g. R&D on effective measures. Consult vet or public health officials.

Fig. 2. On-farm food safety system hazard decision tree.

Having conducted the detailed hazard analysis for both the overall livestock sector and the process steps within a pig production enterprise, the HACCP team then determined the steps which could be designated as Critical Control Points (CCPs). A decision tree was used for this task (Fig. 3). To better link with the approach used here, the decision tree was modied from the one included in the Codex HACCP guidelines (Codex Alimentarius Commission, 2003b) by inserting an intervention step. An intervention step is one which has been specically included to control an identied hazard. It does not refer to a preventive measure applied at a process step undertaken for another purpose. 2.7. Remaining HACCP principles applied to CCPs The HACCP team applied the remaining Codex HACCP principles as specied by the Codex Alimentarius Commission (2003b).  Step 8/PRINCIPLE 3 e Critical limits must be specied and validated if possible for each CCP. In some cases, more than one critical limit may be required for a particular step since they could be performance based or absolute criteria.  Step 9/PRINCIPLE 4 e Monitoring is the scheduled measurement or observation of a CCP relative to its critical limits. The monitoring procedures must be able to detect loss of control at the CCP. The monitoring should provide the information in time to make adjustments to ensure control of the process (in order to prevent occurrences outside the critical limits). The amount of monitoring therefore must ensure that the CCP is in control.  Step 10/PRINCIPLE 5 e Corrective actions were developed for each CCP to deal with deviations when they occur. The action must ensure that the CCP has been brought under control and must also consider disposal of the product. These should be documented in the HACCP record keeping e the documentation requirements for CCPs will be described in industry standards once developed.  Step 11/PRINCIPLE 6 e To determine whether the actual HACCP system is working correctly, verication procedures must be established. These could include auditing/review of procedures and in some cases tests including random sampling and analysis. The frequency of verication should be enough to ensure that the HACCP system is working effectively.  Step 12/PRINCIPLE 7 e Accurate and efcient record keeping is essential to application of a HACCP system. Documentation examples include the hazard analysis, all the reference documents used in the risk assessment, CCP determination and critical limit determination. Record keeping examples include deviations and corrective action reports. The application of these principles can be seen in Tables 6 and 7.

Also listed in Pointon and Horchner (2010) are occupational zoonoses associated with pork production. These are included to clarify that evidence is lacking to classify them as foodborne hazards. Consequently, they are not considered further in this food safety HACCP plan development. Furthermore, information and data for identied and potential microbial foodborne hazards across the Australian pork supply chain are presented in Pointon and Horchner (2010). An internationally accepted severity rating system was used for these porkassociated hazards (ICMSF, 2002). 2.5.2. Hazards at the industry level An assessment of the pork industry was conducted to determine if there were food safety hazards and related threats to market access that should be controlled by the livestock sector therefore providing justication for control of the hazard in on-farm food safety programs. A decision tree developed specically for this purpose was used to determine whether: identied hazards were a threat to market access; identied hazards were a food safety risk; animals are a recognised source of the hazard in meat; and whether there are control measures available on-farm to prevent, eliminate or reduce the hazard to an acceptable level (Fig. 2; Horchner et al., 2006). Each of these hazards identied and classied as being relevant from the risk-based prole (Pointon & Horchner, 2010) were assessed using the on-farm hazard decision tree (Fig. 2) to determine whether it should be included in on-farm food safety programs. The results of the assessment are included in Table 2. Hazards which resulted in a Do not include in on-farm QA program recommendation in Fig. 2 were seen to be either effectively controlled elsewhere in the supply chain, not required to be controlled, or may require further research to understand and/or control the hazard. 2.5.3. Hazard analysis at the enterprise level As described above, an enterprise level hazard analysis was then conducted as applicable to a typical pig production enterprise (Table 3) for each of the process steps and activities in Fig. 1 and

Table 2 Hazard identication step for pork industry overall. Hazard 1. Biological 1.1 Microbiological Salmonella Q1. Food safety market access issue? Y/N Q2. Hazard causing foodborne illness? Y/N Q3. Animals are the source of the hazard in meat? Q4. Effective measures on-farm? Y/N Q5. Effectively controlled elsewhere? Y/N Consider further in on-farm food safety program Y/N

1678

Pig

Singapore

Yersinia enterocolitica Toxoplasma gondii Listeria monocytogenes Campylobacter jejuni/coli Clostridium perfringens E. coli O157:H 7 (EHEC) Indicator bacteriaa e.g. Generic E. coli; Total Viable Counts (TVC); Coliforms Staphylococcus aureus Anti-microbial resistant bacteria Clostridium difcile Salmonella Typhimurium DT104 1.2 Gross carcase lesions Abscess Arthritis Avian Tuberculosis Gross abnormalities Taenia solium (Pork measles)

N N N N N N Y

Y Y Y Y Y Y N

Pig Pig Processing environment Pig Pig/Processing Pig (Beef) Pig/Processing

N N

Y Y

Processing Pig

N N?

N Y

Y Pig

Y Processing controls in place but reduce incidence if pre-slaughter hygiene measures adopted by producers. Slaughter pigs between 6-24 h off feed e but effectiveness and benet: cost of on-farm control is questionable N Y Processing controls in place Y exclude cats, rodents, birds N req additional research for processed sow meats N Y Processing controls N Y Processing controls N Y Processing controls N N req additional research N Y Processing controls in place but made easier if clean livestock N Y Processing controls in place Y Good Agricultural Practice N (GAP) and Good Veterinary Practice (GVP) N N req additional data Y Swill feeding ban Y Processing controls in place Y Animal health program GVP Y for Erysipelas but other causes endemic Y segregate from birds Y Animal health program GVP Y personnel hygiene, ban on using contaminated water Y production away from aquatic intermediate hosts Y Swill feeding ban Y Inspection Y Inspection Y Inspection Y Inspection Y Inspection to detect but more effectively controlled on-farm Y Inspection

Y identify time off last feed

N Y P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688 N N N N N

N Y

N Y

N N N N N?

Y (Salmonella) N N N Y

Pig Pig Pig Pig Pig

Y N N N Y

Spirometra erinacei (Sparganosis) Trichinella spiralis Hydatids Mycotoxins (Ochratoxin A) Pyrrolizidine alkaloids 2. Physical Foreign bodies (needles, maintenance)

N Y N N N

Y Y N Y Y

Pig (outdoor and feral) Pig Pig (feral) Pig Pig

Y (outdoor prod) Y Y Y Y

Y testing of exported feral pork Y worm farm dogs Y Inspection Y feed programs and codes Y stock feed, ingredient, storage, and bedding controls of practice Y stock feed, ingredient, Y feed programs and codes storage, controls of practice

Pig

Y Good Husbandry Practices, Y inspection may detect; but Y includes notication of needle more efcient to control onbreakage farm

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

1679

3. Results 3.1. Determination of hazards for possible control on-farm The hazards for further consideration were identied (Table 2) by using the on-farm hazard decision tree (Fig. 2). These decisions were informed by technical data and information reviewed by Pointon and Horchner (2010). The public health and trade signicance of these were analysed using the risk rating tool (Table 4) and results reported in Table 3. The availability and effectiveness of control measures were reported where the hazard was considered signicant. 3.2. Identied critical control points Using the information from Table 3 in the CCP Decision Tree (Fig. 3), the HACCP team identied process steps that were CCPs covering two main types of hazards, namely the physical hazard of non-recovered broken needles and chemical residues above the MRL. A summary of decisions taken regarding CCP determination for an on-farm food safety program for a pig production enterprise is provided in Table 5. The identied CCPs were:  Selection and sale of breeding stock (broken needles, chemical residues greater than MRL)  Breeding (non-recovered broken needles)  Farrowing (non-recovered broken needles)  Growing stock (non-recovered broken needles)  Inputs (grain treatments, sanitisers, pest control, herbicides for chemical residues above MRL)  Sale pigs (pre-loading check for ag/vet chemical residues above MRL) The application of the remaining HACCP principles in relation to the CCPs is detailed in Table 6. For each hazard the application of simple Good Agricultural Practice (GAP) is effective in minimising the occurrence of the hazard (Table 5). A summary of food safety verication systems used on-farm and during processing, together with surveillance at the industry and public health level, is provided in Table 7. 3.3. Identied control points A number of Control Points was identied for a range of hazards and potential hazards which were found to be managed through the application of GAP (Table 3). The application of GAP applies to appropriate use of veterinary chemicals in animals, on pastures and in feedstuffs fed to animals as well as physical hazards such as broken needles. In relation to the identied microbiological hazards, there are numerous GAP controls to prevent and/or reduce their risk and these controls would also need to be a feature of any on-farm food safety program (Table 2). Examples of GAP include:  Feed withholding to minimise ingesta spillage and carcase contamination with Salmonella at evisceration;  Exclusion of cats, rodents and birds to minimise T. gondii infection;  Judicious use of antibiotics to minimise anti-microbial resistance;  Strict observance of swill feeding biosecurity protocols to prevent introduction of Salmonella Typhimurium DT104 and Trichinella spiralis from contaminated food by-products;

Y N N

Y N

Y Y

Y Y Y Y

Y Porcine Somatotropin N (PST) registered Y Controlled use of Ag & Vet N chemicals. Property risk assessment Y as above N Y Controlled use of Ag & Vet N chemicals. Y as above N Y as above N Y as above N Y as above N

Y as above Y (feeds) N contam. from unrelated industries Y GAP

N N N

Y GAP
a b

N Improves hygiene at carcase dressing but not a Critical Control Point (CCP) for microbial hazards arising on-farm. Not at levels found in meat (National Residue Survey 1996e2008). New routine monitoring.
c

Pig Pig Pig Pig

Nb

Nb

Nb Nb

Nb Y Nb Nb

Y Nb Nc

Pig Pig Pig

Pig

Pig

Pig Pig

Pig N Y

Y Y Y Y

Y Y Y

Y Y

Synthetic pyrethroids Anti-microbial residues Anthelmintics Non-steroidal antiinammatory drugs (NSAIDs) -agonists (Ractopamine only) Heavy metals Dioxins

3. Chemical Hormones

Pest control chemicals e insects, rodents, vermin Sanitisers, disinfectants etc

Organophosphates Macrolytic lactones

Organochlorines

Pig

Table 3 Hazard analysis, control measures and justication for control measures. Process step Associated activity Husbandry differences Hazard Hazard analysis Likelihood Selection of breeding stock Select breeders e identication, traceability, vaccination, oestrus stimulation Purchase breeders e biosecurity, identication, traceability Natural/induced oestrus Severitya Signicance Control measures Justication for control measures

1680

Salmonella

Serious

No effective control measure at this time National Vendor Declaration (NVD) NVD

Broken needles Chemical residues greater than Maximum Residue Limit (MRL) Acclimatisation e vaccination, environ exp., animal treatments Transfer breeders Vaccination

3 3

Serious Serious

Y Y

Quarantine practiced, but not effective against Salmonella NVD contains declaration of status NVD contains declaration of status

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Breeding

On-farm or sale Non-recovered broken needles 3 Serious Y Standard Operating Procedure (SOP) e identify pig to point of slaughter SOP e identify pig to point of slaughter Enables removal at slaughter

Health treatments Mating Gestation Cull breeders Natural, AI Sale

Non-recovered broken needles Sparganosis Salmonella Abscess (with Salmonella) Non-recovered broken needles Pens, stalls, outdoor, shelters, bedding systems Continuous/batch Ear notch, tattoo, Radio Frequency Identication (RFID) Non-recovered broken needles Non-recovered broken needles Crates/outdoor

Serious

Enables removal at slaughter

4 (outdoor only) 2 4 3

Moderate Serious Serious Serious

N Y N Y

Feed to slaughter interval SOP e identify pig to point of slaughter

optimise dressing hygiene Enables removal at slaughter

Health treatments Housing Farrowing Production system Teeth/tails/identication

Wean/foster Vaccination Health treatments Housing Castration Transfer stock Production system

3 3

Serious Serious

Y Y

SOP e identify pig to point of slaughter SOP e identify pig to point of slaughter

Enables removal at slaughter Enables removal at slaughter

Growing Stock

Continuous/batch

Salmonella

Serious

Vaccination Housing Health treatments Transport preparation Transport Tattoo, brand Off-site, pens, shed, shelters, bedding systems, outdoor

Non-recovered broken needles Sparganosis Non-recovered broken needles

3 4 (outdoor only) 3

Serious Moderate Serious

Y N Y

Good Agricultural Practice (GAP) e Limited effectiveness SOP e identify pig to point of slaughter

Some GAP counterproductive Enables removal at slaughter

SOP e identify pig to point of slaughter

Enables removal at slaughter

Inputs

Feed Grain Protein meals Supplements

Home mixed, pelleted, byproducts

Salmonella

2 (protein meals only)

Serious

Protein meals meet Aust rendering standard Vendor Declarations e Grain production and storage Vendor Declarations e Grain harvesting Good Veterinary Practice (GVP) e minimised and targeted anti-microbial usage SOP e Compliance with labels or authorised off-label use

Minimises major pathway

Mycotoxins

Serious

Minimises toxin production Minimises contamination Limits emergence of resistant bacteria

Pyrrolizidine alkaloids Water Ag/vet chemicals Mains, bore, dam, river Animal health Salmonella Anti-microbial Resistance (AMR)

3 4 3

Severe 1B Serious Serious

Y N Y

Animal health, grain treatments

Chemical residues above MRL

Serious (Trade)

Bedding Semen Premises/Facilities (Pests)

Sanitisers, pest control, herbicides Straw, rice hulls, sawdust Prior land use e outdoor, pest exposure, chemical drift, foreign bodies

Chemical residues above MRL Salmonella Salmonella

3 4 3

Serious (Trade) Serious Serious

Y N Y

Compliance with label use.

Labels or instructions for off-label use are comprehensive enough to achieve MRLs Minimise inadvertent exposure

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Toxoplasma gondii

Severe 1B

Foreign bodies

Serious

Pest control programs, limit pest access to feed and bedding Pest control programs, limit pest access to feed and bedding Maintain facilities and clean up after maintenance

Reduces pathway of contamination Reduces pathway of contamination Reduces pathway of contamination

Outputs

Pre-requisite programs Personnel Treatments Efuent disposal

Personal hygiene Recycling mulch from shelters

Pork measles Salmonella

5 3

Moderate Serious

N Y Composting GAP Heat from composting reduces organism to acceptable level.

Sale pigs

Dead pig disposal Sale of breeders Selection of pigs for sale

Salmonella Abscess (with Salmonella) Ag/vet chemical residues above MRL, foreign bodies

2 4 3

Serious Serious Serious (Trade)

Y N Y

Last feed to slaughter interval Check animal status records

Optimise dressing hygiene Prevent noncompliant animals entering the food chain

Pre-loading check

Severity e ICMSF (2002).

1681

1682 Table 4 Risk rating criteria (Anonymous 2005; ICMSF 2002). Likelihood Severity

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Q 1 Was there a hazard at this process step? YES Q 2 Do validated control measure(s) exist for the identified hazard at this step? NO YES Is control necessary at this step or activity? NO Q 3 Is it an intervention step?* Not a CCP YES NO Not a CCP

Severe 1Aa Severe 1Bb Seriousc (Trade) Moderated Negligible 1e 2f 3g 4h 5i Y Y Y Y N Y Y Y N N Y Y Y N N Y Y N N N N N N N N

Modify step, control measure or product

Y Yes; N No. a Severe hazard for general population: life threatening or substantial chronic sequelae or long duration. b Severe for restricted populations: life threatening. c Incapacitating but not life threatening; sequelae infrequent; moderate duration. d Not usually life threatening; no sequelae; normally short duration; symptoms are self limiting; can be severe discomfort. e Common occurrence. f Moderate occurrence. g Uncommon. h Highly unlikely to occur. i Negligible.

NO Q 4 Could contamination with identified hazard(s) occur in excess of acceptable levels or increase to unacceptable level(s)?

YES

YES

NO

 Prevention of Taenia solium (pork measles) through personnel hygiene and ban on using contaminated water;  Location of outdoor production systems away from aquatic intermediate hosts to prevent infestation with Spargana erinacei (Sparganosis). In the context of reducing carcase contamination with foodborne hazards, there appears to be limited value in the application of preventive disease control programs (GAP) to reduce the presence of gross carcase abnormalities at slaughter (Hamilton et al., 2002; Pointon, Hamilton, Kolega, & Hathaway, 2000). Most pathogenspecic abnormalities (Edwards, Christiansen, Johnston, & Mead, 1999) are caused by agents which are not foodborne hazards for humans. Additionally the occurrence of foodborne hazards in non-specic gross abnormalities such as abscesses and arthritis can be predicted to be very low (Pointon et al., 2000). Nevertheless, animal health programs are essential for the wellbeing and welfare
Table 5 Decisions taken on CCPs. Process step Selection of breeding stock Associated activity Purchase breeders e biosecurity, identication, traceability Cull breeders Vaccination Health treatments Vaccination Health treatments Production system Vaccination Health treatments Feed Grain Protein meals Supplements Hazard

Not a CCP

STOP * Critical Control Point for OnFarm Food Safety Scheme

Q 5 Will a subsequent step eliminate identified NO hazard(s) or reduce the likely occurrence to an acceptable level?

NO

YES

Not a CCP

Address through Good Agricultural Practice

* An intervention step is one that has been specifically included to prevent, eliminate the hazard in question or reduce it to an acceptable level.
Fig. 3. Critical control point decision tree.

of livestock. However, apart from the occasional animal condition caused by foodborne zoonoses (Pointon & Horchner, 2010), disease prevention programs have minimal impact on reducing gross abnormalities that are likely to contain low levels of foodborne hazards and as such, fail to meet the requirements of a CCP.

Q1 Y Y Y Y Y Y Y Y Y Y Y Y

Q2 N Y Y Y Y Y Y Y N Y Y Y

Q3 N N N N N N N N N N N N

Q4 e Y Y Y Y Y Y Y e Y Y Y

Q5 e N N Y N N N N e N N Y

CCP N Y (CCP 1) Y (CCP 1) N (GAP) Y (CCP 2) Y (CCP 2) Y (CCP 2) Y (CCP 2) N (GAP) Y (CCP 2) Y (CCP 2) N (GAP)

Breeding

Farrowing Growing Stock

Inputs

Salmonella Broken needles Chemical residues greater than MRL Salmonella Non-recovered broken needles Non-recovered broken needles Non-recovered broken needles Non-recovered broken needles Salmonella Non-recovered broken needles Non-recovered broken needles Salmonella

Ag/vet chemicals Animal health Grain treatments Sanitisers, pest control, herbicides Premises/Facilities (pests)

Outputs Sale pigs

Efuent disposal Selection of pigs for sale Pre-loading check

Mycotoxins Pyrrolizidine alkaloids AMR Chemical residues above MRL Chemical residues above MRL Chemical residues above MRL Salmonella Toxoplasma gondii Foreign bodies Salmonella Salmonella Ag/vet chemical residues above MRL Foreign bodies

Y Y Y Y Y Y Y Y Y Y Y Y Y

Y Y N Y Y Y Y Y Y Y Y Y Y

N N N N N N N N N N N N

Y Y Y Y Y Y Y Y Y Y Y Y

Y Y Y N N Y Y Y Y Y N Y

N (GAP) N (GAP) N (GVP) N (GAP) Y (CCP 3) Y (CCP 3) N (GAP) N (GAP) N (GAP) N (GAP) N (GAP) Y (CCP 4) N (GAP)

Table 6 HACCP table identifying CCPs and application of remaining HACCP principles. Process step & activity Selection of breeding stock Purchase breeders e biosecurity, identication, traceability CCP No. Hazard CCP1 Foreign bodies e Nonrecovered broken needles Control measure Critical limits Monitoring procedure Corrective action Verication methods Regular review of on-farm procedures, records and competency of operators. Feedback from processors.

Identify pig/s with foreign 100% identication of body to enable animals with foreign notication bodies to customers and inspection and removal at slaughter Standard Operating Procedure (SOP) e Compliance with labels or authorised off-label use 100% of animals released within Withholding Period (WHP) and Export Slaughter Interval (ESI) if applicable

CCP1

Chemical residues > MRL

100% check of records of Exclude non-conforming affected animals recorded for animals. identication and subsequent notication Investigate the cause of break down and alter procedure if ineffective. Re-train operators. 100% monitoring of release of Exclude non-conforming animals in accordance with animals from consignment or treatment and exposure notify customer of correct records status Review procedure for identication and traceability, assessment and selection

Regular review of on-farm procedures, records and competency of operators

Breeding

CCP2

Non-recovered broken needles

Vaccination and health treatments

Identify pig/s with foreign 100% identication of body to enable animals with foreign notication to customers bodies and inspection and removal at slaughter

100% check of records of affected animals recorded for identication and subsequent notication

Farrowing

CCP2

Non-recovered broken needles

Vaccination and health treatments

Identify pig/s with foreign 100% identication of animals with foreign body to enable notication to customers bodies and inspection and removal at slaughter

100% check of records of affected animals recorded for identication and subsequent notication

Growing stock

CCP2

Non-recovered broken needles

Vaccination and health treatments

Identify pig/s with foreign 100% identication of body to enable animals with foreign notication to customers bodies and inspection and removal at slaughter

100% check of records of affected animals recorded for identication and subsequent notication

Inputs

CCP3

Chemical residues > Maximum Residue Limit (MRL)

SOP e Compliance with labels or authorised offlabel use

100% of animals released within WHP and ESI if applicable

100% monitoring of release of animals in accordance with treatment and exposure records

Grain treatments

Feedback from processors National Residue Survey (NRS) results and follow up by state departments Producer independent testing as required. Exclude non-conforming Regular review of on-farm animals. procedures and competency of operators. Investigate the cause of break Feedback from processors. down and alter procedure if ineffective. Re-train operators. Exclude non-conforming Regular review of on-farm animals. procedures and competency of operators. Investigate the cause of break Feedback from processors. down and alter procedure if ineffective. Re-train operators. Exclude non-conforming Regular review of on-farm animals. procedures and competency of operators. Investigate the cause of break Feedback from processors. down and alter procedure if ineffective. Re-train operators. Exclude non-conforming Regular review of on-farm animals from consignment or procedures, records and notify customer of correct competency of operators status Review procedure for feeding, Feedback from processors identication and traceability, National Residue Survey assessment and selection results and follow up by state departments Producer independent testing as required. (continued on next page)

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688 1683

1684

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Exclude non-conforming Regular review of on-farm animals from consignment or procedures, records and notify customer of correct competency of operators status Review procedure for Feedback from processors chemical use National Residue Survey results and follow up by state departments Producer independent testing as required. 100% monitoring of treatment Exclude non-conforming Regular review of on-farm and exposure records animals from consignment or procedures, records and notify customer of correct competency of operators status Review procedure for Feedback from processors identication and traceability, National Residue Survey assessment and selection results and follow up by state departments Producer independent testing as required. 100% check of records of Exclude non-conforming Regular review of on-farm affected animals recorded for animals. procedures, records and identication and notication competency of operators. Investigate the cause of break Feedback from processors. down and alter procedure if ineffective. Re-train operators. 100% of animals released with known status Determine the treatment status of animals prior to release CCP4 Pre-loading check Sale pigs Foreign bodies e Nonrecovered broken needles Ag/Vet chemicals > MRL Notication of customer (abattoir or subsequent farm) of identied pig/s to enable inspection and removal at slaughter 100% notication of animals with foreign bodies

Monitoring procedure

100% monitoring of procedures and chemical use records (if appropriate)

In relation to Salmonella spp. that occur widely in lymph nodes of pigs (Pointon et al., 2000) effective measures are not validated onfarm and processing controls elsewhere (e.g. knife sterilisation) are designed to minimise contamination that may arise (Table 2). Further summary comments based on data and information from Australia (Pointon & Horchner, 2010) are provided in Table 2 to give guidance as to the scientic rationale for the HACCP teams consideration of the full range of hazards using the hazard decision tree in Fig. 2. The treatment of Question 5 in the CCP Decision Tree (Fig. 3), regarding subsequent steps as a means of controlling hazards is always difcult for enterprises early in the food chain, chiey because the product is not yet in the form that will be consumed. There is always debate about later steps being more effective for control. With most chemical hazards, this question is easier because there are no subsequent steps able to control the hazard. For microbial and many physical hazards, the question is more difcult. The approach taken by the HACCP team was to consider the controls already in place and mandated by regulations later in the chain. For example, controls in the Australian Standard for Hygienic Production and Transportation of Meat and Meat Products for Human Consumption (AS4696-2007) apply to all meat processing enterprises and therefore, all products are subject to these control measures (Anonymous, 2007). 4. Discussion The application of HACCP for the pig production sector, advocated in the Code of Practice for the Hygienic Production of Meat (Codex Alimentarius Commission, 2005), has been used to provide technical underpinning for on-farm food safety programs based on a major review of the available scientic information. HACCP guidelines from the Codex Alimentarius Commission (2003b) were followed. The approach (Horchner et al., 2006) resulted in the identication of CCPs for selected physical and chemical residue hazards and their implementation on-farm is scientically justiable, resulting in a positive outcome with regard to minimising food safety and trade-related food safety hazards. The approach avoids identication of unsubstantiated or seemingly arbitrary measures that have undetermined importance to the ultimate goal of reducing consumer risk. Nevertheless, a resulting HACCP-based food safety program based on the ndings of this project would provide pig producers with a framework that is understandable and realistic to apply, two essential elements that underpin successful implementation and compliance by industry. While it appears difcult to satisfy HACCP principles for other hazards onfarm, especially microbial hazards, account was taken of the existence or otherwise of effective downstream control/reduction measures and the need for effective intervention type CCPs rather than prevention or reduction type control steps. Downstream controls include HACCP requirements at abattoirs and in the production of processed meats (Anonymous, 2007) and systems to verify their effectiveness (AQIS; ESAM; NRS; OzFoodNet; NEPSS Table 7). As hazards change, new production systems are developed and as new science becomes available then this justication for onfarm food safety programs should be reviewed and modied accordingly (Codex Alimentarius Commission, 2003b). In the guidelines for application of the HACCP system (Codex Alimentarius Commission, 2005) it is noted that it is important when applying HACCP to be exible where appropriate, given the context of the application taking into account the nature and the size of the operation. When considering the application of HACCP for the pig production sector the HACCP team recognised the large number of individual enterprises that would be required to

Corrective action Critical limits Control measure CCP No. Hazard Table 6 (continued ) Process step & activity

Verication methods

Sanitisers, pest control, herbicides CCP3

Chemical residues > MRL

SOP e Compliance with labels or authorised offlabel use

100% compliance with label directions or authorised off-label use

Table 7 Overall/summary food safety verication plan for non-CCP controls and surveillance. Verication Activity Examples of verication activities On-Farm Activities Internal Audit program Description & reference documents/procedures for this activity Frequency Overall responsibility for this activity Relevant Records

Audits e Australian Pork Industry Quality (APIQ)/PigPass

Commodity Vendor declarations National Vendor Declaration

Internal audits conducted using internal auditors covering all aspects of the HACCP program over a year. All WIs, SOPs and control measures reviewed External audit by an accredited auditor covering the scope of the farms HACCP-based quality system. Evidence of compliance with the APIQ/PigPass standards via the HACCP plan and associated SOPs are assessed through the examination of records, staff interview and farm based observations Vendor declaration provided by the supplier of the commodity. Process covered in farm quality system, SOPs etc Vendor declaration covering the relevant food safety hazards. Process will be covered in farm quality system, SOPs etc

As per Internal Audit Schedule Annual

Farm Manager/Owner

Internal audit reports, farm records, customer complaints, laboratory reports, processor feedback Farm records including Internal & External Audit reports, treatments records, staff training records, APIQ/PigPass records, laboratory reports, corrective action requests Vendor declaration APIQ, PigPass

APIQ/PigPass auditor

As dictated by the farm quality system. All sale lots

Farm Manager/Owner Farm Manager/Owner Livestock Personnel at processing plants External Auditor

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Customer Group External Audits

Second party audit of the farm and its quality system by the purchaser of the pigs

As required

Water testing e voluntary testing

Covered by the farm quality system and associated procedures

As required

Reviews of Corrective Action records and product dispositions

Internal auditing process undertaken by a designated staff member in which examination of the relevant farm documents/ records and a possible review of quality system SOPs occurs. These records are also examined as part of the annual external audit

As dictated by the quality system

Farm Manager/Owner Quality Assurance Manager at processing plants Farm Manager/Owner

Farm records including Internal & External Audit reports, treatments records, staff training records, APIQ/PigPass records, laboratory reports, corrective action requests Laboratory reports

Internal & External Audit reports, CARs, customer complaints, processor feedback.

Review of QA records

Internal auditing process undertaken by a designated staff member in which examination of the relevant farm documents/ records and a possible review of quality system SOPs occurs. These records are also examined as part of the annual external audit

As dictated by the quality system

Quality Assurance Manager at processing plants Farm Manager/Owner

Farm records including Internal & External Audit reports, treatments records, staff training records, PigPass records laboratory reports, corrective action requests

Review of training records

Internal auditing process undertaken by a designated staff member in which examination of the relevant farm documents/ records and a possible review of quality system SOPs occurs. These records are also examined as part of the annual external audit

As dictated by the quality system

Quality Assurance Manager at processing plants Farm Manager/Owner

Internal & External Audit reports, staff training records, training certicates

Investigations of residue violations ChemCert records Post farm gate activities TruckCare Abattoir Food Safety Regulator External Audits Ante and Post mortem inspection E coli and Salmonella Monitoring (ESAM)

State primary industry personnel investigations and documentation Auditing process undertaken by scheme administrators

As required As dictated by the quality system As dictated by the quality system Monthly SA 6 monthly Continuous Sample per:

Quality Assurance Manager at processing plants State primary industry jurisdictions Scheme Administrators e auditors engaged Scheme Administrators e auditors engaged Australian Quarantine & Inspection Service (AQIS) State Meat Hygiene Authorities AQIS Company QA AQIS

Audit reports

Auditing process undertaken by scheme administrators Export abattoirs are audited by Area Technical Manager Domestic audit frequency varies by state: All ante/post mortem inspection carried out on all pigs/carcases May be by AQIS or company inspector AQIS (for export) oversees routine carcase sampling for E. coli and Salmonella spp.

Audit reports AQIS State Meat Hygiene Authorities AQIS e total condemnations. Companies report to producers (ad hoc) Data held by AQIS. (continued on next page)

1685

1686

Table 7 (continued ) Verication Activity Description & reference documents/procedures for this activity Frequency 1,000 pigs for E. coli 5,000 pigs for Salmonella Sampling rate laid down by AQIS programme (300 samples/residue to give a 95% at 1% prev) Overall responsibility for this activity Relevant Records

National Residue Survey (NRS) testing

Sampling at slaughter to detect violative levels of chemicals and contaminants

NRS e Department of Agriculture, Fisheries and Forestry (DAFF)

NRS report

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

Customer audits Port of entry testing Overall industry activities Regulation of Agricultural and Veterinary (APVMA) chemical use OzFoodNet National Enteric Pathogen Surveillance Scheme (NEPSS) Anti-microbial Resistance Survey (AMR) Monitoring of hazards in foods

At moment violative results from APIQ members are sent to APL (AQIS having discussions with APL to expand to cover all detections) As specied by customers e.g. supermarkets Testing of imported Australian pork at port of entry for Salmonella and chemical residues Risk assessments conducted by APVMA to establish usage patterns, safety and withholding period standards Quarterly and Annual reports of notiable foodborne disease incidence nationally Annual reports of national foodborne hazard laboratory identication conrmations National survey of foods for presence and prevalence of AMR in key foods (includes pork) National survey of foods for presence and prevalence of residues and contaminants National regulations to prevent transmission of exotic diseases into piggeries Review of the on-farm HACCP plans by HACCP team

Reports to state jurisdictions of violations

As required by customer As per country requirements Continuous

Commercial auditors Import country trade jurisdictions Australian Pesticide and Veterinary Medicines Authority (APVMA) Dept Health and Ageing Dept Health and Ageing AQIS

www.apvma.gov.au

Continuous Continuous

www.ozfoodnet.org.au NEPSS, 2007

Ad hoc Periodic

Dept Health and Ageing Food Standards Australia New Zealand (FSANZ) State animal health jurisdictions APL

Barlow & Gobius, 2008 www.foodstandards.gov.au/scienceandeducation/ monitoringandsurveillance/ australiantotaldiets1914.cfm State animal health jurisdictions HACCP plan This report Risk proles Risk-based hazard review (Appendix 2) and supporting reference documents

Swill Feeding regulation compliance Re-assessment of HACCP plan & components Review of Validation documents

Continuous Periodic

Periodic review of intrinsic hazards and reports such as this that underpin the industry level and on-farm program controls

Periodic

Australian Pork Limited (APL)

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688

1687

replicate such a detailed hazard analysis as produced in this paper. It also recognised that the amount and level of technical information required to conduct a comprehensive hazard analysis at the individual enterprise level would be beyond the skills and resources of most livestock producers. Consequently, the HACCP team modied the approach by initially applying the HACCP principles at an industry level to derive appropriate on-farm food safety control measures applicable at the enterprise level for pig production. To ensure coverage of all possible scenarios on-farm that might contribute to risk, a detailed risk-based prole and intrinsic hazards summary of the pork industry was prepared and used to inform the approach at the industry level (Pointon & Horchner, 2010). This identication of hazards at the industry level was repeated at the level of the individual livestock enterprise. This through chainenterprise level approach enabled a simpler process for determining critical control points (Figs. 2 and 3; Tables 2 and 5). By applying this HACCP-based approach it was determined that the application of a set of Good Agricultural Practices (GAP) onfarm (Tables 2 and 3) are effective in ensuring low risk and if maintained by industry, will keep it low. It is, therefore, concluded that requirements of on-farm food safety programs may not warrant full HACCP plans for all hazard types at the individual pig production enterprise level as long as appropriate GAP is in place. It is noted that this approach can only be taken in industries/countries such as Australia that have mandated control post-primary production and can validate the outcomes by surveillance and public health data (Table 7). While there is no CCP for the control of Salmonella on-farm (Goldbach & Alban, 2006; Mousing et al., 1997) an example of the control of risk through GAP is withdrawing feed for a period before slaughter to minimise intestinal trauma and ingesta spillage at evisceration (Pointon & Horchner, 2010). Combined with Good Manufacturing Practice (Anonymous, 2007) and emerging interventions such as hot-water decontamination of carcases at slaughter (Hamilton et al., 2010; Hurd et al., 2008) it is left to multiple hurdles of effective GAP/GMP controls to ensure a prevention and minimisation approach is taken to keep hazard levels acceptably low (Coates et al., 1997; Hamilton et al., 2007). In terms of verication of the effectiveness of these GAP/GMP measures in Australia to minimise risk from microbial hazards, national surveillance programs provide on-going public health information to detect emerging problems (OzFoodNet; NEPSS). These programs are supported by targeted risk-based surveys for specic hazards, for example the recent national survey for the prevalence of anti-microbial resistant organisms associated with food, including pork (Barlow & Gobius, 2008). In relation to ensuring risk from agricultural and veterinary chemical residues is negligible, the industry and government maintains an integrated system of chemical registration (APVMA), licencing of veterinarians to prescribe the use of medications with written instructions on treatment and withholding periods, national vendor declarations for livestock (NVD), farmer training on use of chemicals (e.g. Chemcert), routine residue monitoring in food products (NRS, 1996e2008) and monitoring consumer exposure (FSANZ, 2008). These controls provide for the safe use of otherwise potentially hazardous chemicals (Tables 2 and 3). While the consumer risk from chemical residues at levels found in the Australian pork is negligible (NRS, 1996e2008) the signicance of these potential chemical hazards and food safety threat to market access is profound in terms of market maintenance. Levels are low because GAP controls are effective. When problems occur it is usually found that these controls are not maintained (Bill Salter APL, pers. comm.). Consequently the breadth of any on-farm food safety program or system should encompass chemical usage to ensure market access risk is managed. In this context the hazard

decision tree (Fig. 2) was developed to assist the HACCP team identify the risks and controls that need to be covered in an on-farm food safety program. As a consequence of such an approach, any on-farm food safety program will still need verication at the industry level (NRS, 1996e2008) to support market access as well as be supplemented with compliance audits at the enterprise level. Development of on-farm auditing arrangements might similarly adopt a risk-based approach. Examples of current risk-based programs in the Australian pig industry that should be considered in development of auditing arrangements include the Australian Pork Industry Quality Program (APIQ and PigPass). Further examples of verication programs summarised in Table 7 show that Australia has extensive reviews of the outcomes of its on-farm food safety systems with the livestock sector as well as downstream. Based on the results of the process, the HACCP team concluded that it is appropriate to implement on-farm food safety systems based on Good Agricultural Practices (GAP) which address control measures for the CCPs and non-CCPs identied for pig production. The up-to-date review of hazards and control measures provided here are a key technical underpinning for the hazard analysis at the industry and enterprise levels (Pointon & Horchner, 2010). This should be reviewed on an on-going basis to ensure the resulting onfarm food safety program reects contemporary knowledge of risk. This would provide an appropriate level of protection for consumers as well as protect Australias trade risks from food safety related issues.

Acknowledgements The authors gratefully acknowledge the contributions of the HACCP team assembled for this project: David Hamilton, Kim Nairn, Paul Pattison, Pat Mitchell and Jo Slade. The authors also thank Australian Pork Limited for their nancial support. References
Anonymous. (2005). Australian standard AS/NZS 4360 Risk management and handbook HB 436 Risk management guidelines e Companion to AS/NZS 4360. Standards Australia. Anonymous. (2007). Hygienic production and transportation of meat and meat products for human consumption. Food Regulation Standing Committee Technical Report Series 3. AS 4696:2007. Standards Australia. APVMA(Australian Pesticides and Veterinary Medicines Authority). Australian Government. www.apvma.gov.au. AQIS. (2009). Requirements for AQIS certication regarding chemical residues. AQIS Meat Notice 2009/06. Australian Bureau of Statistics. (2008). Value of agricultural commodities produced. 2006e07. Australia. http://www.ausstats.abs.gov.au/ausstats/ subscriber.nsf/0/E8E50264CDED1AF1CA2574960012927A/$File/75030_200607.pdf. Australian Bureau of Statistics. (2010). Value of principal agricultural commodities produced. Australia Preliminary. 2008e2009. http://www.ausstats.abs.gov.au/ Ausstats/subscriber.nsf/0/4CA4802CED3D2ADACA2576C400130742/$File/ 75010_2008-09.pdf. Australian Pork Ltd. (2008). Australian Pig Annual 2006e2008. Deakin West, ACT, Australia: APL. Barlow, R., & Gobius, K. (2008). Pilot survey for antimicrobial resistant (AMR) bacteria in Australian food. Canon Hill, Queensland: Australian Government Department of Health and Ageing, Food Science Australia. Coates, K. J., Kolega, V., Barton, M., Barlow, S., Groves, M., Hamilton, D., et al. (1997). Level of food-borne pathogens on Australian pig meat and carcasses. In Manipulating Pig Production VI (pp. 148). Canberra, Australia: Australasian Pig Science Association. Codex Alimentarius Commission. (2002). Proposed draft principles and guidelines for the conduct of microbiological risk management. CX/FH 03/7. Joint FAO/WHO Food Standards Programme. Orlando, Florida: Codex Committee on Food Hygiene CX/FH 03/7. Codex Alimentarius Commission. (2003a). Proposed draft process by which the Codex Committee on Food Hygiene could undertake its work in microbiological risk assessment/risk management. CX/FH 03/6. Joint FAO/WHO Food Standards Programme. Orlando, Florida: Codex Committee on Food Hygiene CX/FH 03/6.

1688

P.M. Horchner, A.M. Pointon / Food Control 22 (2011) 1674e1688 Horchner, P. M., Brett, D., Gormley, B., Jenson, I., & Pointon, A. M. (2006). HACCPbased approach to the derivation of an on-farm food safety program for the Australian red meat industry. Food Control, 17, 497e510. Hurd, H. S., Ene, C., Srensen, L., Wachmann, H., Corns, S. M., Bryden, K. M., et al. (2008). Risk-based analysis of the Danish pork Salmonella program. Past and future Risk Analysis, 28, 341e351. ICMSF. (2002). Microorganisms in foods: 7 microbiological testing in food safety management. New York: Kluwer Academic/Plenum Publishers. Mousing, J., Thode Jensen, P., Halgaard, C., Bager, F., Feld, N., Nielsen, B., et al. (1997). Nation-wide Salmonella enterica surveillance and control in Danish slaughter swine herds. Preventive Veterinary Medicine, 29, 247e261. NEPSS(National Enteric Pathogens Surveillance Scheme). (2007). Non-human annual report 2006. Microbiological Diagnostic Unit, The University of Melbourne. NRS. (1996e2008). National residue survey annual reports. Canberra, Australia: Australian Government Department of Agriculture, Fisheries and Forestry. OECD/WHO. (2003). Foodborne disease in OECD countries. Present state and economic costs. Paris: Organisation for Economic Co-operation and Development, World Health Organization. OzFoodNet. A health network to enhance the surveillance of foodborne diseases in Australia. www.ozfoodnet.gov.au. Pointon, A., & Horchner, P. (2010). Food safety risk-based prole of pork production in Australia. Technical evidence to support an on-farm HACCP scheme. Australian Pork Ltd, ISBN 978-0-646-52134-3. Pointon, A. M., Hamilton, D., Kolega, V., & Hathaway, S. (2000). Risk assessment of organoleptic postmortem inspection procedures for pigs. The Veterinary Record, 146, 124e131.

Codex Alimentarius Commission. (2003b). Recommended international code of practice general principles of food hygiene (CAC/RCP 1e1969, Rev. 4-2003). Codex Alimentarius Commission. Codex Alimentarius Commission. (2005). Code of hygienic practice for meat. CAC/RCP 58e2005. Codex Alimentarius Commission. Edwards, D. S., Christiansen, K. H., Johnston, A. M., & Mead, G. C. (1999). Determination of farm-level risk factors for abnormalities observed during post-mortem meat inspection of lambs: a feasibility study. Epidemiology and Infection, 123, 109e119. FSANZ(Food Standards Australia New Zealand). The Australian Total Diet Study (ATDS). www.foodstandards.gov.au/scienceandeducation/monitoringandsurveillance/ australiantotaldiets1914.cfm. FSANZ. (2008). The 22nd Australian total diet survey. Canberra: FSANZ. Goldbach, S. G., & Alban, L. (2006). A cost-benet analysis of Salmonella-control strategies in Danish pork production. Preventive Veterinary Medicine, 77, 1e14. Hamilton, D., Holds, G., Lorimer, M., Kidd, C., Slade, J., & Pointon, A. (2010). Decontamination of pork carcases with hot water or acidied sodium chlorite e a comparison in two Australian abattoirs. Zoonosis and Public Health, 57, 16e22. Hamilton, D. R., Gallas, P., Lyall, L., Lester, S., McOrist, S., Hathaway, S. C., et al. (2002). Risk-based evaluation of post mortem inspection for pigs in Australia. The Veterinary Record, 151, 110e116. Hamilton, D. R., Smith, P..Pointon, A. (2007). National Salmonella and E. coli Monitoring (ESAM) data from Australian pig carcases from 2000 to 2006. In: Proceedings 7th international symposium on the epidemiology & control of foodborne pathogens in pork (pp. 129e132). Verona, Italy.