Food Safety Interventions: Reducing Risk from Farm to Table
By Martin W. Bucknavage, M.Sc., MBA and Catherine N. Cutter, Ph.D.
The term “intervention” is used by the food industry when discussing microbiological control. From the dictionary, “intervene” means “to come in or between so as to hinder or modify.” From a food safety perspective, intervention involves the addition of control measures into a process to reduce, and ultimately, prevent or eliminate food safety risks. Food safety professionals are often tasked with adding interventions to existing processes to provide additional barriers to control potential microbiological hazards where there is chance that measures downstream may not eliminate that hazard. This approach may be accomplished to meet governmental regulations, such as Salmonella reduction in a poultry facility, or to meet performance standards, a company’s internal standards or the standards of their customers.
Pathogens on raw poultry is a good example of where interventions are needed. While the consumer is expected to cook raw poultry to eliminate Salmonella or Campylobacter, we know that correct handling and preparation by the consumer is not foolproof. The consumer may undercook the product or cross-contaminate their counter, cutting board or knife with raw poultry juices. To reduce the risk of foodborne illness at the consumer level, the industry employs many interventions from hatching and grow-out, as well as during transport and processing, to reduce the prevalence of Salmonella and Campylobacter on the raw product. By decreasing the prevalence of pathogens in the live bird and throughout the process, the consumer is less likely to encounter the pathogens if and/or when the product is mishandled.
Interventions can be categorized as to where they occur within the food supply chain. Generally speaking, pre-harvest interventions occur in the field or on the farm, while post-harvest interventions occur as the raw material undergoes processing. In the food manufacturing facility, there are controls placed into the process or at packaging, antimicrobials may be added to the product or into the packaging material and controls can be employed in the sanitation process to reduce the prevalence of food safety hazards in the processing environment. Finally, training can be used as an intervention to educate employees, as well as consumers, on proper handling of food products to reduce the risk of foodborne illness.
The food industry is interested in controlling pathogenic bacteria from “farm to fork” or “gate to plate.” While many pre-harvest interventions for meat and poultry are still in the research phase, there have been promising results for the use of on-farm practices, such as vaccines or competitive exclusion, as a way to eliminate pathogenic bacteria from the environment and intestinal tracts of live animals.
There has been great interest in using practical and cost-effective interventions to reduce the prevalence of pathogens in live animals. For example, heat treatments can reduce pathogen prevalence in contaminated feed. Competitive exclusion is accomplished when healthful bacteria are added to feed so the bacteria out-compete pathogens in the gut of the animal, resulting in a reduction or exclusion of pathogenic microorganisms, such as Salmonella or Escherichia coli O157:H7. The industry also uses environmental controls on the farm or in poultry growing houses, such as enhanced cleaning procedures as an intervention to reduce the incidence of pathogenic bacteria.
In 2009, the U.S. Department of Agriculture (USDA) conditionally granted approval for a vaccine against E. coli O157:H7 for use in cattle. Studies indicated the vaccine reduced the number of cattle that tested positive for the pathogen by 85%. Of the animals that did test positive for the pathogen, the vaccine reportedly eliminated the organism in a large percentage of the animals. The vaccine company estimates that the vaccine will protect as many as 10 million cattle every year (about 1/4 of the country’s annual cattle supply) and will likely cost less than $10 per head.
Another pre-harvest approach involves Good Agricultural Practices, a food safety program that addresses a number of interventions to reduce the risk of pathogens on ready-to-eat (RTE) fruits and vegetables. An important intervention of interest to fruit and vegetable growers is elimination of pathogens from irrigation water. Often, water is pulled from nearby ponds or streams and may contain pathogens. The restriction of cattle near sources of irrigation water is one way to address this contamination risk. Additional concerns include sources of pathogens from natural fauna, farm employees or sources that are further upstream. Further interventions may be needed to treat irrigation water, especially when water will be used on food products before harvest.
Post-harvest interventions are employed to reduce contamination during processing, especially where there is potential for cross-contamination and bacterial proliferation. With cattle, organisms such as E. coli O157:H7 can be associated with the hide, hooves as wel as in the intestinal tract. As the animal is slaughtered, bacterial interventions, such as hide cleaning or bung tying, help reduce the risk of contamination of the carcass.
The use of antimicrobial carcass sprays or steam vacuuming reduces the risk of pathogenic E. coli on the exterior portion of the carcass. When processing trim into ground beef, for example, the application of lactic acid or other antimicrobials prior to grinding may be utilized as an intervention aimed at reducing pathogenic E. coli in the final product.
For poultry, antimicrobial treatments with chlorinated compounds are added to the chill water to reduce the risk of pathogens circulating in the chilling tank water. Whole carcass washes with antimicrobials immediately after leaving the chill tanks and prior to packaging also are used to reduce microbial loads on the final products.
Fruit and vegetable processors use a number of post-harvest interventions. Cooling produce rapidly after the product is picked will not only improve the quality of the product and extend its shelf life, but also will help limit the growth of pathogens picked up in the field. Fruit and vegetables often receive an application of sanitizer solution, such as ozone or chlorinated compounds after the wash step, as an intervention to reduce the risk of pathogens.
Within further processing facilities, interventions can be applied throughout the process, from initial processing of raw materials to the packaging step. The requirement of prior testing of an ingredient, as evidenced by a certificate of analysis, can be considered an intervention step to prevent highly contaminated product from entering the facility.
For the RTE food industry, contamination by Listeria monocytogenes is a major concern. During sanitation, interventions may provide assurance that the processing environment will not be a source of the pathogen on the finished product. The use of an antimicrobial fog and/or misting sanitizer throughout the facility after hours may reduce the prevalence of the pathogen in the environment. Floor foamers also may be used in areas where there is a potential risk of transporting microbes from raw or outside areas to high-risk areas. Additionally, compounds such as ozone can be used to rinse food contact surfaces, thereby reducing environmental pathogens and/or spoilage organisms.
In addition to an effective sanitation program, a number of different in-plant interventions can be applied to control L. monocytogenes. For example, RTE meat and poultry industries can incorporate potassium lactate, sodium diacetate or another approved antimicrobial to the product formulation to restrict pathogen growth during the shelf life of the product. Prior to packaging, RTE products can be sprayed with an antimicrobial to eliminate surface contamination. Treatments, such as heat applied to the packaged product, can inactivate the pathogen on the product surface.
Finally, incorporation and/or slow release of antimicrobial agents into bio-based or vacuum-packaged materials may provide a means of extending the bacterial lag phase, reducing microorganism growth rate and extending the shelf life of some RTE foods.
Training and Education
Food safety training and education of industry personnel and food handlers could be considered an intervention as well. In fact, the World Health Organization has suggested that education and training of food handlers and consumers is the single most important way to prevent foodborne illnesses.
Numerous academic institutions and industry organizations have training tools (e.g., PowerPoint presentations for face-to-face training, online courses, case studies, examples of food safety plans, videos, flip charts, booklets, Cooperative Extension publications, etc.) that can be used to educate the food industry workforce, including non-English-speaking employees. These low-tech training devices have proved to be valuable tools in training farm employees as well as food handlers in processing establishments with limited financial resources or employees who have difficulty understanding food safety training delivered in English.
Consumer education is another intervention that can be used to reduce foodborne illness. Cooking or handling instructions that take into account the wide range of possible consumer actions can go a long way to reduce the risk of foodborne illness originating in the home. There are many food safety campaigns by different organizations, including the U.S. Food and Drug Administration (FDA), USDA, Cooperative Extension and consumer groups that all emphasize the importance of training consumers in safe food handling and preparation.
Interventions as Part of an HACCP System?
Can an intervention be a Critical Control Point (CCP) in a Hazard Analysis and Critical Control Points (HACCP) plan? In some instances, yes, especially when it is used to control a hazard that is likely to occur and when the intervention is specific to the process. Interventions also can be part of a prerequisite program such as those seen with sanitation.
For example, floor foamers may be put in place to control the spread of L. monocytogenes from one area to another in an RTE establishment. In this case, the foamer, or intervention, is an additional component of the sanitation program that keeps low-level hazards from becoming “likely to occur” hazards. Designated as a part of a prerequisite program, such a sanitation intervention may cross multiple product lines. However, if a failure occurs (e.g., foamers are empty), it is not likely to result in a food safety hazard or concern.
However, if we consider the use of antimicrobial sprays on meat animal carcasses, the spray should be considered a CCP. Research has previously demonstrated that the use of such a spray will result in a reduction in the number of microorganisms, such as pathogenic E. coli, on the surface of that product. As a CCP, employees must ensure that all carcasses receive the spray treatment. Furthermore, the concentration of the antimicrobial in the spray must be verified, as well as the volume of spray applied to each carcass.
Incorporating an Intervention
A facility may decide to incorporate an intervention as a means of adding safety to the product. Rarely, if ever, does an intervention come with no cost to the establishment. At a minimum, there are labor costs associated with the day-to-day operation of the intervention. Before an intervention is undertaken, it is important to ask some basic questions. Is it feasible to add the intervention to the facility? Is the intervention affordable? And most importantly, will the intervention provide the level of performance (safety) expected?
When discussing the feasibility of incorporating an intervention, both process and facility design must be considered. Is there adequate space to accommodate the intervention? Will the intervention negatively affect product flow? Are adequate inputs available to meet demands, such as electrical and water usage, personnel and even technical expertise?
The total capital costs associated with the implementation of an intervention often go beyond just the cost of the equipment. In addition to equipment costs, there are costs for making changes to the facility and process. In many cases, these can be as much, if not more, than the equipment costs. Then there are the variable costs of running the equipment. If the intervention is an antimicrobial spray, what are the costs of the chemical, as well as the costs of the electricity and water to run it? In addition, there are ongoing costs for labor, both direct and indirect, maintenance of the intervention and verification testing for meeting HACCP regulations. Additionally, there are start-up costs, including validation testing and additional labor costs for these initial steps. Downtime and product loss encountered during the trial phase also may add to the initial costs. In some instances, these costs may not be recognized.
If, after consideration of the above issues, one finds implementation and operation are not cost-prohibitive, the intervention may be required to meet performance requirements. In many cases, a performance standard is established, either internally by a customer or by a governmental agency. For example, a regulatory agency may require that a finished product be free of a pathogen (zero tolerance). In another instance, performance of an intervention may be determined by how well the microbial or bacterial load is reduced; also known as “log reduction.” For a one-log reduction, 90% of the chosen bacteria will be reduced; for a two-log reduction, 99% of the bacteria will be reduced, 99.9% for a three-log reduction, etc. Therefore, a five-log reduction would result in 99.99% of the bacteria being reduced.
In the case of E. coli O157:H7, USDA considers it an adulterant in ground beef, meaning that not even one cell of the pathogen can be present in the product. For the processor of fresh juice, FDA has set a requirement that an intervention (e.g., pasteurization) affect a five-log reduction of E. coli O157:H7 to be considered effective.
Once the level of a performance standard is determined, validation of the given intervention is conducted. As part of an initial validation, available scientific data (e.g., peer-reviewed journal publications) and regulatory documents may be evaluated to determine whether there is solid scientific data to back the intervention performance claim. It must be determined whether there is any regulatory information for the intervention in question. If so, does the information support the facts in that it will meet the standard of performance for the type of product being processed and the type of process being used?
Too often, establishments adopt an intervention when there is little or no scientific evidence to support the performance. In some cases, it must be determined whether the parameters, resulting data and supporting information match processing conditions used in the facility. If and when insufficient data exist, challenge studies may be conducted. These studies should be designed to reflect the actual conditions used within the establishment. Studies also should address the pathogen(s) of concern, either through the use of the organism, or if not feasible, an appropriate surrogate or indictor organism can be substituted. Under laboratory conditions employed for a challenge study, the product is inoculated with a high level of organisms and then the inoculated product is exposed to the intervention. After the intervention, the product is analyzed using microbiological analyses to determine the number of surviving organisms. A log reduction is calculated by subtracting the log of the number after the intervention from the log of the number before the intervention. Of course, a statistically determined number of samples must be analyzed to provide validity to the process.
Challenge studies can be expensive and for the small establishment, they may be cost-prohibitive. Another drawback is that the laboratory conditions used in the challenge study can be different from those used in the processing facility. In some cases, establishments may be presented with challenge studies funded by the intervention equipment manufacturer or studies conducted for a competitor. Under these circumstances, validity of the intervention and/or data may come into question. Clearly, these issues should be addressed before moving forward on the purchase of the intervention equipment.
If at all possible, in-plant trials of interventions should be performed prior to implementation and preferably before purchase of any equipment. The collection and evaluation of in-plant data for an intervention is an important part of the validation process. Data should be collected on process parameters to ensure the intervention is working as it was designed. It also is important to conduct microbiological analyses to support the process. While it is unlikely that product will be inoculated with pathogens for testing in the plant, microbiological testing can be completed using generic bacteria such as aerobic plate counts (APCs), Enterobacteriaceae, non-pathogenic (generic) E. coli and/or coliforms as performance indicators. This testing should yield results consistent with performance of the intervention. For example, if the goal is to affect a three-log reduction of pathogenic E. coli, then there should be no generic E. coli present in the product after the intervention is applied to an experimentally inoculated product. Additionally, a concurrent reduction in both APCs, Enterobacteriaceae and coliforms should be observed. If desired, testing for naturally occurring pathogens can be performed by conducting microbiological analyses that either enumerate or detect the presence/absence of the organism of interest.
In cases where findings are questionable or do not result in a sufficient microbial reduction, then one may question the validity of the intervention and whether it is worth the cost. It is important to review all intervention parameters in relation to the process to ensure that there have been no changes. In one instance, intervention under-performance occurred when the product flow through the intervention was much higher than the intervention was capable of handling. In this particular case, the product was not adequately exposed to the intervention, and thus, the reduction in APCs was less than expected. Additionally, management under-reported the production level in order to save money, resulting in a less robust spray intervention.
Once an intervention is implemented, more in-plant data should be collected on process parameters, and microbiological verification testing should be performed periodically. An implementation phase that incorporates a statistically valid testing schedule should be developed to substantiate the performance of the intervention. As this point in the process, testing should take into account varying production parameters such as different shifts, seasonality and variations in raw material.
Over time, ongoing verification activities should continue to ensure that the intervention is operating as it should. Verification activities may include microbiological analyses as well as the monitoring of process parameters. It is important for these data to be analyzed and negative trends investigated as soon as possible.
Interventions can be added to a process as a means to reduce the risk of potential hazards, most notably, pathogenic microorganisms. They can be added throughout the product flow from pre-harvest though processing of a finished product. However, it is important to make sure the intervention is worth the cost in terms of achieving required performance levels. Validation by evaluation of supporting documentation, as well as in-plant microbiological testing, should be conducted, especially when large financial resources are allocated.
Martin Bucknavage, M.Sc., MBA is senior food safety extension associate at the Department of Food Science, Pennsylvania State University.
Catherine N. Cutter, Ph.D. is associate professor and food safety extension specialist at the Department of Food Science, Pennsylvania State University.
1. www.fsis.usda.gov/PDF/Reducing_Ecoli_ Shedding_In_Cattle_0510.pdf.
2. Cox, J.M. and A. Pavic. 2009. Advances in Enteropathogen Control in Poultry Production. J Appl Microbiol 108:745–755.
3. www.defendingfoodsafety.com/2009/03/articles/food-safety-news/usda-conditionally-approves-new-e-coli-o157h7-vaccine-for-cattle/. Accessed 12/12/10.