concept as applied to some chemical, physical and microbiological contaminants of milk on dairy

Veterinary Quarterly ISSN: 0165-2176 (Print) 1875-5941 (Online) Journal homepage: http://www.tandfonline.com/loi/tveq20 The Hazard analysis critical control point's (HACCP) concept as applied to some chemical, physical and microbiological contaminants of milk on dairy farms. A prototype J.J. Lievaart, J.P.T.M. Noordhuizen, E. van Beek, C. van der Beek, A. van Risp, J. Schenkel & J. van Veersen To cite this article: J.J. Lievaart, J.P.T.M. Noordhuizen, E. van Beek, C. van der Beek, A. van Risp, J. Schenkel & J. van Veersen (2005) The Hazard analysis critical control point's (HACCP) concept as applied to some chemical, physical and microbiological contaminants of milk on dairy farms. A prototype, Veterinary Quarterly, 27:1, 21-29, DOI: 10.1080/01652176.2005.9695183 To link to this article: https://doi.org/10.1080/01652176.2005.9695183 Copyright Taylor and Francis Group, LLC Published online: 01 Nov 2011. Submit your article to this journal Article views: 2309 View related articles Citing articles: 9 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalinformation?journalcode=tveq20

21 Veterinary Quarterly 2005; 27(1): 21-29 The Hazard Analysis Critical Control Point's (HACCP) concept as applied to some chemical, physical and microbiological contaminants of milk on dairy farms. A prototype J.J. Lievaart*, J.P.T.M. Noordhuizen, E. van Beek, C. van der Beek, A. van Risp, J. Schenkel and J..van Veersen Faculty of Veterinary Medicine, Utrecht University, The Netherlands SUMMARY Quality management on dairy farms becomes more and more important regarding the different areas of animal health, animal welfare and food safety. Monitoring animals, farm conditions and farm records can be extended with risk identification and risk management. The hazard analysis critical control point's system is useful as an on farm strategy to control the product as well as the production process on the areas of animal health, animal welfare and food safety. This article deals in detail with the question how to develop a qualitative method where risk can be defined as an interaction between probability and impact. Two parts of the production process (milk harvest and treatment of cows) where used as an example how to apply the hazard analysis critical control point's system on chemical, physical and microbiological contaminants of milk. Not just only by summarizing the different critical checkpoints for each area but also by giving them a precise judgement of probability and impact. Keywords: Animal health; Animal welfare; Cattle; Food safety; HACCP; Hazard Analysis Critical Control Point; Milk contaminants; Public health; Quality management; Risk identification; Risk management; Zoonoses. *Corresponding author Address: Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, Yalelaan Z 3584 CL Utrecht, The Netherlands E-mail: j.j.lievaart@vet.uu.nl

22 INTRODUCTION Quality assurance is becoming increasingly important in the primary livestock production sector. This has been provoked, not in the least, by food safety failures like BSE, VTEC, dioxin contamination of animal feedstuff, which have extensively been addressed in the media in the recent years. Consumers have an increasing impact on the way in which animal products are being produced, but also on the manner of keeping animals, e.g. regarding welfare issues. Quality in animal production, like in other more industrial branches, refers to the product, e.g. milk, and or to the production process, e.g. feeding, milk harvesting or calf rearing. Quality control of milk has been performed since many years. Classical checkpoints are, for example, butterfat and milk protein content, bacterial counts, somatic cell counts and antibiotic residues (8,9). With regard to the production process on a dairy farm one may consider issues like animal health, animal welfare and public health as features of that production process (14). Cows "pass" several steps in that process and at every step some activities, e.g. feeding, milking, animal health care, take place. Animal health care is classically centred around diagnosing and treating diseased animals. Since in the 1970's programmes for herd fertility have been introduced (4), other herd health and production management programmes have been developed and introduced into dairy practice (2). Core elements in such programmes are goal setting, protocols for execution of the programme, problem analysis, planning of actions and evaluation. Monitoring animals, farm conditions, management and available on-farm records are the key activities during regularly planned farm visits. Such monitoring is meant to obtain early signals about animals which are possibly deviating from expected performance or about pending herd problems. The ultimate aim is to install risk identification and risk management plans on the dairy farm with a focus on prevention (15). The Hazard Analysis Critical Control Point's (HACCP) concept has been proposed as an on-farm strategy to address both product and production process related quality issues regarding animal health, welfare and public health (3,14). Key elements of the HACCP-concept are hazard identification, risk identification and analysis, dfinition of critical control points on the farm, documentation and verification of health status and actions to maintain or improve that status and quality risk management. It might be that application of the HACCP-concept to the dairy farm can integrate operational herd health, production management and quality assurance issues. To the knowledge of the authors, elaborations of such applications on the dairy farm have so far not been published. Therefore, the objectives of this paper are to present in more detail the most relevant features of the HACCP-concept, and illustrate its application by addressing two steps in the production process where zoonotic diseases have a large impact: the milk harvesting and the treatment of cows. MATHERIALS AND METHODS Features of the HACCP-concept HACCP has been formally adopted by the European Union to assure quality issues in different kinds of food producing industries. The following principles (table 1) have been given to characterise HACCP (3) Table 1. The 7 principles of HACCP 1. Give a detailed description of the production process, e.g. by using flow diagrams. 2. Identify and evaluate potential hazards and risks related to the hazards during the process of production. 3. Determine critical control points in the production process where such risks can be identified and controlled. 4. Specify when at a critical control point the process step is under control, by setting standards, criteria, tolerances. 5. Design an on-farm monitoring system involving all critical control points to check whether all specifications are being met. 6. Determine the corrective measures for situations where critical control points are beyond their control limits. 7. Verify the HACCP-plan by using additional information and or actions. From table 1 it can be read that such a HACCPplan can only be implemented when the dairy farmer has adopted the proper mentality and attitude toward quality in its broadest sense. It has been stated before that HACCP means "formalizing and structuring what the truly good farmer is doing anyway" (19). HACCP is highly farm-specific,

23 focused on self-management, and rather low graded in labour-intensity, while at the same time relatively few documents are needed, when compared to other quality concepts like ISO 9000. For clarity reasons it is instructive to have the production process diagram available in hard copy. It facilitates the risk assessment in a given production step by its specific definition; it shows the reader where certain conditions are interrelated. An example is given in figure 1. Two production steps from this figure, milk harvesting and the treatment of cows, will be highlighted in detail. Specify Grassland Management.. Purchase / Harvest 4 ( Materials r \.i Pregnant Heifers 4 \. } Storage Rearing / Replacement (Thughage, Concentrates, \. Byproducts Feeding.-- Water.1 Water supply Air --) Semen & Medicinal drugs _i 0. Climatise barn Treatment & Care Air r i " Female calves \.. / - ( Cleaning & Desinfection materials } 4' Calving r--.. Male Calves \ _ I Milk Harvesting --. Colostrum, Pen Milk } I. Milk Storage Delivery Cows ) )(Waste / Dead cattle) c -. Manure \, E Milk ) Figure 1. Overview of the most important steps in the dairy farm production process

24 Critical control points, CCP's, are the pivot of the HACCP-plan. They refer to the hazard of concern, should be measurable in the production process and ascertain that control of the hazard is feasible. It should also be coupled with standards and tolerance levels, while control should lead to elimination or reduction of the risk. An example of a CCP is the temperature of the rinsing water for the milking machine with the objective of cleaning the milking equipment without leaving risks for contamination of the system: standard level 80 Celsius with tolerance of 2 degrees. Often these CCP's are physical in nature, for which standards and tolerances can be defined, while in other cases such CCPs are more biological or managerial in nature for which strict standards and tolerances have not or hardly been defined. An example approach for the latter may be the CCP for purchasing cattle: do not purchase cattle with an unknown origin and unknown health status; if one desires to purchase a cow ask for a health certificate or do a pre-test before entry into the herd. The corrective action will be when the cow appears to be infected with disease X, either to treat the animal or to cull her. RESULTS The HACCP-concept and the process steps of milk harvesting and cow treatment In figures 2 and 3 the outlines of these production steps have been put into more detail. As one can read from these figures, the respective elements within each process step are structured more formally. Milking machine ready Collect animals Entry into milking parlour / Udder towels i -\ Concentrates._ 2 ( Dip or spray J Cow identification y } 0. Udder preparation --\ Attach milking cluster Collect milk and transport Post milking treatment ( I, Deliverable milk OE -N \..1 - Non-deliverable milk } Let animals leave milking parlour Cleaning water Clean the milking parlour Waste water OM Clean and disinfect milking machine Rinsing water _ -1 Cleaning and disinfecting \.. agents 1 Figure 2 The production process step of milk harvesting

25 O. ( Sick cow } Culling 4-- Herd Herd formularium and/ or consult the vet r- Healthy cow -00 Preventive treatment V Record sick cows and treatment 41111---- r ---, Medicinal drugs. V Withdrawal period Treatment result Figure 3. The production process step regarding treatment and care of cattle The first question then is which the hazards of concern are. Hazards may be physical, chemical (cleaning and disinfection liquids, antibiotic residues) or microbiological in nature. The latter refer among others to zoonotic agents. The risk analysis step in the HACCP-procedure addresses probabilities and effects. Risk can be defined as the outcome of the interaction between probability and impact (R = P*I). Risks can be quantified by applying formal quantitative epidemiological principles and methods, yielding odds ratios from observational-analytic field studies (11,15). Another, semi-quantitative approach to rank risk factors in order of relevance has been proposed by Steenkamp (21) and Horst, Huirne and Dijkhuizen (7). These authors used adaptive conjoint analysis and area-experts to obtain this ranking order. Relatively little effort has been put into identifying risk factors and quantifying their contribution to the occurrence of a disorder. Therefore, a third approach is addressed here to a more qualitative method. The probability that a hazard may occur can be classified by grade 1 (very low probability), grade 2 (low probability), grade 3 (high probability), grade 4 (very high probability). In the same way the effect of a hazard may be classified by grade 1 (very little effect), grade 2 (little effect), grade 3 (moderate effect), grade 4 (large effect), grade 5 (very large effect). These grades are assigned to each risk factor of estimated significance to the hazard occurrence, by which a relative weighing is obtained. With regard to milk-borne zoonotic diseases in The Netherlands Listeria monocytogenes, Salmonella spp., Staphylococcus aureus, Escherichia coli 0157:H7 appear to play a role of significance on dairy farms based on their prevalence surveys (18). Some of these agents are detectable in milk after faecal contamination, others due to (sub)clinical mastitis. Risk factors of these diseases have been presented by Fourichon, Seegers, Beaudeau and Bareille (5), Baljer and Wieler (1), Hassan (6), Zadoks et al. (22), Sanaa, Putrel, Menard and Serieys (20), Nielsen, Tegtmeier, Andersen, Gronbeak and Anderson (12), Pak (17) and Jayarao and Henning (10). Some of these risk factors form part of a more general good farming practice code, a code of attitude and mentality, while others are more specific for a given disease.

26 The outcome of a risk estimate for physical and chemical hazards of milk in the part-process of milk harvesting and cow treatment is presented in table 2. In this table also the process step and the CCP are identified and and the preventive measures are described. A comparable risk estimate can be given for microbiological hazards in the process step of milk harvesting (table 3). At the same time this table comprises the CCP-identification, the control measures and the prevention measures. It is obvious that both tables are relate. After all they originate from the same production process. From the tables it can also be seen that based on the risk estimates a certain priority setting in actions can take place. Control and preventive measures are given as illustration how it may look like. Finally, it has been shown that there is a mixture of the proper attitude (GFP), the critical management aspects (CMP) and the critical control points (CCP) from the HACCP-plan in the field. The next step in the procedure is the design and implementation of a HACCP control plan for a dairy farm. This involves the different process steps, the kind of control point, the monitoring, the instruction leaflets, the corrective measures (for control and prevention) and the record keeping. An example of part of a HACCP-controlplan for the process step "cow treatment" is presented in table 4. The example given in table 4 shows that implementing a HACCP-plan is basically the structuring and formalizing of what a truly good farmer is doing anyway. It is indeed part of a general good management that only appropriate drugs are being administered to cows, that advised withdrawal periods are complied with and that a proper dosage is applied in order to avoid the risk of drug residues in milk. The appropriate identification of treated cows is relevant, again for avoiding drug residues in milk and moreover for avoiding treating the wrong cow. The latter therefore is a CCP. The drug advisory plan of the Table 2. Outcome of risk estimate regarding the physical and chemical hazards of milk CodeC/F Hazard Process step WeighingRisk. M =milking P and I Yes/No T=treatment P I P*I CCP/ CMP GFP Prevention/control measure (farmer focused) Ti. C Wrong drug T 2 2 4 No CMP Proper diagnosis Check label Drug advisory plan See instruction leaflet T2 C Residue in meat T 1 2 2 No CMP Withdrawal time Drug advisory plan See instruction leaflet T3 C Wrong dosage T 2 2 4 No CMP Check proper dosage Use drug advisory plan T4 C Drug >shelflife T 2 1 2 No GFP Check shelflife See instruction leaflet.. T5 C Cow's ID fails during withdrawal T 2 3 6 Yes CCP See instruction leaflet T6 F Needle broken T 1 1 1 No GFP Inject with care M1 C During residues in milk M 2 3 6 Yes CCP Good cow ID Withdrawal time Stick to indications See instruction leaflet T=treatment step in process. M=milk harvesting step in process. C=chemical. F=physical. GFP=good farming practice code. CCP=critical control point. CMP=critical management point or point of attention. P=probability. 1=impact.

27 Table 3. Risk estimate of microbiological hazards during milk harvesting Hazard Process step Risk estimate P I P*I Risk Yes/No CCP/CMP/ Prevention/control GFP measures Responsible person Milk with T, M 1 2 2 No GFP Certificate free of Farmer Brucella Brucella abortus abortus High hygiene level Milk with T, M 1 3 3 No GFP Certificate free of Farmer Mycobacterium Mycobacterium bovis bovis Milk with T, M, BT 2 4 8 Yes CCP Hygiene at milking Farmer Listeria Infection cows separate monocytogenes Reduce infection spread Milk with T, M, F 1 3 3 Yes CMP Hygiene at milking Farmer Salmonella Increase health status dublin/typhimurium Milk with T, M, C 2 2 4 Yes CMP Hygiene at milking Farmer Corynebacterium jejuni Milk with Staphylococcus aureus T, M 3 2 6 Yes CCP Cows with infection to be separated; reduce infection spread (see drug advisory plan & instruction leaflet) Farmer Milk with T, M, C 2 3 6 Yes CCP Hygiene at milking Farmer Escherichia coli 0157117 Milk with T, M, C 1 2 2 No. GFP Hygiene at milking Farmer Yersinia enterocolitica T=treatment step in process. M=milk harvesting step. F=feeding step. C=barn climate step. BT=bulk milk tank. GFP=good farming practice code. CCP=critical control point. CMP=critical management point or point of attention. Table 4. Example of part of a HACCP-control-plan of a dairy farmer for chemical and physical hazards in the process step of cow treatment (see also figure 1 for total production process) Item nr. CCP/ Standard Monitoring CMP Tolerance How Freq Who Instruction (prevention) Corrective measures Records Ti CMP Use only Check At use Farmer proper drug label "Use of drugs" Use proper drugs Evaluate other drugs Consult the vet "Drug record" T2 CMP No residues Check At delivery Farmer "Use of drugs" Respect the withdrawal drugs "Drug delivery" periods "Drug record" 13 CMP in Check At use Farmer "Use of drugs" Adjust dosage "Drug record" D.A.P. syringe 14 CCP Cow ID Visual At drug use Farmer "Use of drugs" Mark the cow "Drug record" No tolerance T=cow treatment step in the process. CMP=critical management point. CCP=critical control point. D.A.P.=drug advisory plan of the veterinarian for the herd.

28 veterinarian appears to be highly valuable as support to the quality control by the farmer. Such a plan should comprise the appropriate drugs for given medical indications and providing the proper dosages and proper application routes. Instruction leaflets, like use of drugs and delivery of drugs, are meant for the farmer to consult in case of doubt. The "use of drugs" instruction leaflet will comprise items like: make sure the right diagnosis is set; use the drug according to the drug advisory plan; take the right drug and check the label for first use date, expiring date, batch number; dosage according to drug advisory plan; cow identification means; record keeping; application of a test on drug residues in case of doubt about delivery of milk (CCP) and archive the instruction leaflets of medicines. Many of these issues are addressed in formally planned veterinary herd health and production management programmes for dairy farms (2). These programmes focus on herd health goals, execution of a programme, for example on udder health management, evaluation of performance by assessing performance figures, and adjustment of the programme when needed. DISCUSSION From the forenamed issues it appears to be feasible to design a HACCP-based plan for dairy farms in the area of chemical, physical and microbiological hazards regarding milk quality. Comparing this approach with the one proposed for dairy herd health and production management (2), it appears that both can be integrated into one programme rather easily. This will stimulate the dairy farmer to address operational herd health and production management as well as the more strategic quality risk management (14). At the same time it shows that there is no threat to the dairy farmer that for the implementation of different programmes he has to deal with different organizations and different people. Not in the least because it can be expected that in addition to the forenamed issues, others are likely to come to the farm. Examples are the EU directive 97-12 and the food safety regulation 178-2002. The former addressing the compulsory farm inspection for animal health, welfare and food safety, the latter addressing the liability and responsibility of farms producing (raw) foods of animal origin. It could be expected as well that this will impose HACCP-like programmes on farms, including farm inspections and demonstration of the status regarding animal health and welfare as well as food safety and public health. In that respect the license to produce and to market farm products is at stake. Consumer demands in the four areas would be met more easily than before. The introduction of HACCP-like programmes will not run smoothly. First, farmers are in general not willing to change their routine practices. Second, they will demand to know what the ultimate goals are that we are heading for. In that respect it would be worthwhile to show them the developments that take place in rapid pace with regard to food safety objectives and consumer protection, for which animal health, animal welfare and production process control are supportive issues. Production process control, risk management, prevention and liability are key elements in this approach. Pivotal in the development of HACCP-like plans is the determination of objective parameters for inter-farm comparison and rewarding strategies. Statistical process control tools, such as capability indices, should be developed for that purpose for the dairy farm, as has been proposed by Niza-Ribeiro and Noordhuizen (13). Thirdly, farmers should be convinced of the fact that quality control may help in reducing quality failure costs and improving market retain instead of increasing profits and expanding market segments. Finally, if we would be able to clarify to farmers that, next to reduction of quality failures costs, such programmes would be helpful in the early warning of impending problems and hereby they would save disease losses by implementing risk management strategies, they may get convinced that indeed they are producing for society and not to the dairy factory itself. They might loose their license to produce and to market their products. In that respect, the impact of public opinion may be stronger in the end than that of rational production based on economics alone. For the veterinarian the question is whether they would take up the challenge to expand their activities from the classical animal and the herd health advisory practice to veterinary quality consultancy in its broadest sense.

29 REFERENCES 1. Baljer G and Wieler L. Animals as a source of infections for humans: diseases caused by EHEC. Deutsche Tierärztliche Wochenschrift 1999; 106: 339-343. 2. Brand A, Noordhuizen JPTM and Schukken YH. Herd health and production management in dairy practice, 2nd edition, 1997. Wageningen Press Publications. Wageningen, The Netherlands. 3. Cullor JS. HACCP (hazard analysis critical control points): is it coming to the dairy? Journal of Dairy Science 1997; 80: 3449-3452. 4. Esslemont RJ, Bailie JH and Cooper MJ. Fertility management in dairy cattle. Collins Publications, 1985, London. 5. Fourichon C, Seegers H, Beaudeau F and Bareille N. Critical control points analysis for udder health management in dairy herds. 47th Annual Meeting of the European Association of Animal Production, EAAP, Lillehammer, Norway, 26-29 August 1996. 6. Hassan L. Farm management and milking practices associated with the presence of Listeria monocytogenes in New York dairy herds. Preventive Veterinary Medicine 2001; 51: 63-73. 7. Horst HS, Huirne RBM and Dijkhuizen AA. Eliciting the relative importance of risk factors concerning contagious animal diseases using conjoint analysis: a preliminary survey report. Preventive Veterinary Medicine 1996; 27: 183-195. 8. IDF Methods for assessing the bacteriological quality of raw milk from the farm. Bulletin no. 256/1991, International Dairy Federation, 1991, IDF, Brussels, Belgium. 9. IDF Detection and confirmation of inhibitors in milk and milk products. Bulletin no. 258/1991, 2nd edition, International Dairy Federation, 1991, IDF, Brussels, Belgium. 10. Jayarao BM and Henning DR. Prevalence of food borne pathogens in bulk tank milk. Journal of Dairy Science 2001; 84: 2157-2162. 11. Martin SW, Meek AH and Willenberg P. Veterinary epidemiology: principles and methods, 2nd edition, 1995. IOWA State University Press, AMES. 12. Nielsen EM, Tegtmeier C, Andersen HJ, Gronbeak C and 1 Andersen JS. Influence of age, sex and herd characteristics on the occurrence of verocytotoxin producing Escherichia coli 0157 in Danish dairy farms. Veterinary Microbiology 2002; 88: 245-257. 13. Niza Ribeiro J, Noordhuizen JPTM and Menezes J.C. Capability index. A statistical process control tool in aid to udder health control in dairy herds. Journal of dairy science 2004; 87: 2459-2467. 14. Noordhuizen JPTM and Welpelo HJ. Sustainable improvement of animal health care by systematic quality risk management according to the HACCP concept. Veterinary Quarterly 1996; 18: 121-126. 15. Noordhuizen JPTM and Wentink H. Developments in veterinary herd health programs on dairy farm: on review. Veterinary Quarterly 2001; 23: 162-169. 16. Noordhuizen JPTM, Frankena K, Thrusfield MV and Graat EAM. Application of quantitative methods in veterinary epidemiology. Wageningen Press Publications, 2001, Wageningen, The Netherlands. 17. Pak S. Risk factors for Listeria monocytogenes contamination of dairy products in Switzerland, 1990-1999. Preventive Veterinary Medicine 2002: 53: 55-65. 18. RIVM. Year report 2002. Zoonoses and zoonotic agents in human, food, animals and feed in the Netherlands in 2001. Inspectorate for Health Protection and Veterinary Public Health, Bilthoven 2002, The Netherlands 19. Ryan D. Three HACCP based programmes for quality management in cattle in Australia. Dairy Extension 1997, NSW, Australia. 20. Sanaa M, Poutrel B, Menard JL and Serieys F. Risk factors associated with contamination of raw milk by Listeria monocytogenes in dairy farms. Journal of Dairy Science 1993: 76: 2891-2898. 21. Steenkamp JBEM. Conjoint measurement in ham quality evaluation. Journal of Agricultural Economics 1987; 38: 473-480. 22. Zadoks RN, Al lore HG, Barkema HW, Sampimon OC, Wellenberg GJ, Giohn YT and Schukken YH. Cow and quarter level risk factors for Streptococcus uberis and Staphylococcus aureus mastitis. Journal of Dairy Science 2001; 84: 2649-2663.