A Dialogue on Pathogen Reduction

The purpose of the meeting was to stimulate thinking and generate new ideas toward enhancement of the Agency’s farm-to-table approach for ensuring the safety of meat, poultry, and egg products. The conference focused on four of the forces that may affect or measure pathogen load—interventions, performance standards, Hazard Analysis Critical Control Points (HACCP), and microbial testing. It consisted of presentations by four panels, each followed by a question-and-answer period. About 200 persons from industry, government, and academia attended the public meeting, one of several that FSIS plans to hold over the next year to strengthen the scientific basis of its regulatory public health programs.

Karen Hulebak, then Senior Advisor for Scientific Affairs, Office of the Administrator, FSIS, opened the conference by thanking the participants for their interest in and willingness to discuss “the substantive scientific issues that are the subject of this meeting and that are so central to the formulation of public health protective policy.” FSIS’s organizing principle for this symposium, she said, was to “generate new ideas for strengthening the effectiveness of HACCP, microbial testing, pathogen reduction measures, interventions, performance standards, inspection, risk assessment, and voluntary measures on the farm. We want to engage in dialogue at this meeting about how pathogen reduction is achieved through such measures.” 

Introduction of Hazards, Farm to Table
Michael Doyle, Regents Professor of Food Microbiology and Director of the Center for Food Safety at the University of Georgia, chaired Panel 1, which set the stage for dialogue by outlining the favorable routes of entry for pathogens at various points on the farm-to-table continuum. Commenting on two case-control studies to identify risk factors associated with harmful microorganisms, Doyle lauded the Centers for Disease Control and Prevention (CDC) and the state health departments for the work that they have done in this area, “giving us some scientifically based guidance and direction to identify those risk factors that are largely responsible in influencing the transmission of foodborne pathogens.”

David Dargatz, Epidemiologist at the Centers for Epidemiology and Animal Health of USDA’s Animal and Plant Health Inspection Service, discussed pathogen prevalence on the farm. He summarized findings from studies of several animal production systems (feedlot cattle, dairy, layers, swine) between 1994 and 2000 conducted by APHIS’s National Animal Health Monitoring System, which gathers information to estimate the prevalence of pathogens and evaluate the risk factors affecting prevalence. In a 1999–2000 study, 100% of feedlots had cattle with one or more samples positive for E. coli O157:H7. This apparent dramatic increase over the 63% of feedlots with positive samples in 1994 is probably due to more sensitive culture methods using immunomagnetic bead separation and more intense sampling in the feedlots. 

Dargatz concluded that pathogens on the farm are mostly present in low prevalence or low numbers and are widely distributed by geography and operation type. So far, across epidemiologic studies, there is inconsistency in identified risk factors for shedding of pathogens. He recommended that data from large studies should be used to generate hypotheses that can be evaluated in more controlled experiments.

--- PAGE BREAK ---

Gary Acuff, Professor of Food Microbiology at Texas A&M University, discussed the steps in the slaughter process for cattle, hogs, and poultry, and the opportunities for pathogen introduction. “The problem is that we can limit some of the contamination but we can’t absolutely assure the absence of the organisms,” he said. 

The primary sources of contamination at slaughter continue to be feces, hide contact, aerosols and sprays, contaminated hands or equipment, and the spilling of bodily fluids, he said. He concluded that if you don’t control the major sources, time spent on controlling the minor sources of contamination is wasted.

Larry Decker, Chief Food Inspector with the New York State Dept. of Agriculture and Markets, discussed processing and post-processing hazards from his perspective. “We’re not finding so many infected workers as we used to in the past,” he said, but infected food handlers remain the number-one problem. Other hazards range from personal hygiene problems to transportation to failure to follow an approved processing schedule, he said. 

He commented that New York State does not allow vacuum packaging of processed meat products at retail because of a concern about Listeria and Clostridium botulinum. A growing problem in many states is custom slaughterers who are becoming “mini-processors,” not only slaughtering the animal or bird for the owner but also making the jerky and dried sausages and in some cases vacuum packaging the products. Most states don’t have much regulatory authority over these businesses yet. 

The hazards are basic, Decker concluded, and so is much of the solution. “All of us need to take the time to educate those who are responsible for handling and processing our meat and poultry products,” he said.

Robert Tauxe, Chief of the Foodborne and Diarrheal Diseases Branch, Division of Bacterial and Mycotic Diseases, with the CDC’s National Center for Infectious Diseases, focused on preparation, consumption, and the chain of transmission. “Each year, an estimated 76 million cases of foodborne illness occur,” he said. “That is one in four of us, one in four Americans.” 

Acknowledging the complexity of preventing hazards from farm to table, Tauxe said, “I need to explain to my public health colleagues sometimes that this isn’t like measles or polio where a simple vaccine is the issue, but that it’s a much more complex partnership.” Variables responsible for complicating the picture include not only the presence of pathogens on raw foods of animal origin, but also the variety of pathogens and their reservoirs, and the several different pathways of transmission—including specific foods, water, direct contact with animals, and direct contact with other humans. He said that although outbreaks garner more attention, “Actually a vastly greater number of illnesses occur as sporadic cases, individual cases, perhaps part of unrecognized disbursed outbreaks.” 

Food preparation adds new variables, Tauxe said. From a review of 2,700 outbreaks between 1993 through 1997 that were reported to CDC, 43 % were associated with restaurants, delis, or other commercial food establishments. The most common contributing factors at the food preparation stage identified in case reports were poor holding temperature (73%), poor personal hygiene (38%), and inadequate cooking (21%). 

A separate review of 1,800 outbreaks in New York State between 1980 and 1995 found that an ill food handler played a role in 23% of cases. “Sick-leave policies in rapid turnover industries, like the fast-food restaurant business, are pretty minimal,” Tauxe said. He suggested that if systematic hand-washing and sick-leave policies were to increase, there might be a measurable association with foodborne illness statistics, just as there is an association between the introduction of HACCP and a decline in foodborne illness rates. 

--- PAGE BREAK ---

Impacts of HACCP Systems and Approaches
Susan Sumner, Professor and Head, Dept. of Food Science and Technology at Virginia Tech, chaired Panel 2, focusing on the impacts of process control programs on pathogen reduction. She noted, “I think there are a couple of things that we do know about HACCP, and how we measure whether it’s successful or not. First, HACCP has to build on the standard operating procedures, standard sanitation operating procedures, and good manufacturing practices, or HACCP is not going to really do anything for us. Second, we have to bring in the aspect of management. We have to make HACCP ‘company culture.’”

Delila Parham, Chief of the Zoonoses Branch in the FSIS Office of Public Health and Science, recapped the history of the Pathogen Reduction/HACCP rule. “Our objective was to build into food processes effective measures for reducing and controlling pathogenic microorganisms,” she said. The rule has three key elements. Each plant must develop and follow a written sanitation plan—the sanitation standard operating procedure. Each plant must develop and follow a HACCP plan for its processes and products, identifying the hazards and establishing ways to prevent and control those hazards. Finally, both plants and FSIS must conduct microbiological testing, and FSIS personnel verify that plants are doing so. 

“The Agency recognizes that the Salmonella performance standards were not based on a risk assessment or a scientifically established public health outcome,” Parham said. “Yet, despite this and despite the limitations of its data, FSIS believes the data show that the contamination levels on raw meat and poultry products have been reduced as a result of the HACCP rule.”

John J. (Jack) Guzewich, Director of Emergency Coordination and Response for the Food and Drug Administration’s Center for Food Safety and Applied Nutrition, asked, “How can we measure the impact of our regulations, whether at the federal, state, or local levels, on public health?” He reviewed the value and the incompleteness of various types of data in answering that question. Such data include disease trends (from both outbreaks and sporadic cases), special epidemiological studies (such as age and risk factors), microbial test results, inspection or compliance findings, consumer behaviors and attitudes. 

He summarized key problems with standard methods used for evaluating impact: lack of baseline data, inconsistent data collection over time, lack of consistent epidemiological and laboratory methods, and outcome data not collected for the purpose of evaluating impact. “One of the problems we have,” he said, “is that an awful lot of the data that we’re using to make conclusions that HACCP implementation has had an impact on public health was collected for legitimate purposes and may provide useful insights, but it wasn’t collected to answer this question.” 

Guzewich also noted that the very success of HACCP implementation “confounds” the difficulty of teasing out relationships between cause and effect. He concluded, “Actually, I think HACCP implementation has had an impact on public health, but not for the reasons that we can measure. It will be very difficult for us to demonstrate that any one intervention has resulted in a specific percentage reduction in foodborne disease in general or any one agent or vehicle. However, HACCP has changed the focus of government, industry, academia, and consumers. HACCP has prompted us to improve surveillance of disease and of agents in foods. The dialogue HACCP has prompted has had a major impact on the changes in how food safety has been addressed in all types of interventions compared to 10 years or so ago. . . . Finally, we can say that FoodNet data show a decrease in some major pathogens at the same time as HACCP and a number of interventions have been implemented, and there’s got to be some linkage there, although we can’t show it statistically.”

--- PAGE BREAK ---

Alejandro Castillo, Associate Professor of Food Microbiology at Texas A&M University, provided a brief overview of HACCP implementation in Latin America. “Costa Rica would like to make HACCP mandatory for every food product,” he said. HACCP is required for seafood in Mexico, for both exports and products consumed domestically. In Argentina and Uruguay, HACCP is required for exported meat and fish, but not for any product sold in the country. 

In general, he said, Latin American countries are focusing on HACCP for export, but are slower in applying the same requirements to domestic product. “Most of the food processors in Latin America see HACCP and food safety as a means to stay in business because even though HACCP is not mandatory by law, many times purchasers require it,” he said. 

What factors are needed to make HACCP more successful in Latin America? “Training, training, and training,” Castillo said. Some company managers believe a short training course is adequate to make them experts, or reject training as too expensive. He said that one way to evaluate the quality of training is to determine if it is accredited. The International HACCP Alliance reviews and endorses training programs and trainers, and, he said, “HACCP training programs that are successful have sent their programs for review to the Alliance.” An accreditation office physically located in Latin America would be helpful, he added. 

Performance Standards and Microbial Testing
Gary Acuff chaired the panel on performance standards and microbial testing. He presented a brief history of the evolution of microbiological performance standards over the past 50 years.

Elise Golan, Senior Economist with USDA’s Economic Research Service, spoke on performance standards and the economics of compliance and innovation. She described three types of safety standards. Target standards, the least intrusive, do not prescribe any specific safety standards for a product or process but impose criminal liability for pre-specified harmful consequences arising from the product. However, she said, target standards are unworkable for food safety because it’s very difficult to draw the line between a foodborne illness and a specific food. Process standards are the most intrusive, because they specify the type of production method and the exact procedures to be used to produce a good. Specifications can be compelling or prohibitive. Milk pasteurization is an example of a food safety process standard. 

“The fact that performance standards specify requirements in terms of results and not production methods has pushed them to the top of economists’ most favored policy tool list,” Golan said. “However, the logic that leads economists to conclude that performance standards encourage efficiency and innovation is built on the premise that the firm is ultimately responsible for the safety of the product,” she added. “Recalcitrant firms not interested in food safety will only have an incentive to be efficient and innovative if violation of performance standards means that the firm will incur real costs and that the firm will ultimately be responsible for rectifying the lapse in safety. If instead the government is responsible for investigating safety lapses and the government is responsible for deciding how safety lapses should be rectified, then the economic logic of safety standards breaks down,” she said. 

Robert Buchanan, Director of the Office of Science in FDA’s Center for Food Safety and Applied Nutrition, spoke on microbial testing for control verification. He described microbiological testing as “one of the most important tools we have for improving the safety of the food supply” and “one of the most poorly understood.” The key, he said, is “to pick the right tool for the job.” To do that, one must understand the goals of different approaches to testing, the basic assumptions that underlie the testing, and the characteristics of the testing program. 

--- PAGE BREAK ---

Within-batch testing, Buchanan said, is used to demonstrate the safety of a single lot of food. It assumes no prior knowledge of the process or food product, yet provides a very detailed snapshot of an operation. However, he cautioned, within-batch testing is effective only within certain ranges of contamination. As a general rule of thumb, he said, if the acceptable defect rate you’re seeking is less than 1%, you should look at measures other than microbial testing because the number of samples that must be collected and analyzed becomes a limiting factor. 

The primary function of between-batch testing is to demonstrate that a food safety system or process is continuing to function as intended, Buchanan said. It assumes an intimate knowledge of the process and prior analysis of process performance and variation. The safety of a specific batch can be assumed if the process has been validated and is in control. 

The flow of microbiological contamination, he said, can be compared to the path of ping pong balls thrown into a river. “If you took a thousand ping pong balls and threw them into a stream, you would not walk upstream to find the ping pong balls; rather, you should find them distributed downstream.” For this reason, if between-batch testing is possible at only one point in the process, end-point testing is best. “However,” he said, “if you’re trying to be proactive and eliminate problems in your process, it is very beneficial to take samples at several points in the process so when you start having problems you can identify where those problems took place and correct them.” 

Regardless of where in the process samples are taken, effective between-batch testing requires enough sampling over time to allow for statistically valid results on which sound conclusions about process control can be based, he said.

Frank Busta, Professor Emeritus, University of Minnesota, said that the requirements for an ideal food safety microbial indicator are that it is easily and rapidly detectable, easily distinguishable from normal flora, present when the pathogen is present, with growth requirements equivalent to those of the pathogen, and absent when the food is free of a pathogen. 

Over the years, a variety of organisms have been used or proposed as indicators. Coliforms are best used as indicators in something that has been processed, but they may or may not indicate fecal contamination, he said. E. coli is regarded as a valuable indicator of fecal contamination of processed products and an indicator of inadequate processing. It is not necessarily a reliable indicator of fecal contamination of raw products or post-processing contamination, because it can and does grow in that environment. “In the past,” Busta said, “we’ve preferred microbial indicators that are not pathogens because of their easier laboratory detection. However, we’re now looking at the advantages pathogens may pose as microbial indicators as more rapid and sensitive detection methods are developed.”

Loren Lange, Acting Associate Deputy Administrator of the FSIS Office of Public Health and Science, spoke on performance standards, statistical sampling, and the process FSIS used to develop the Pathogen Reduction/HACCP rule. 

--- PAGE BREAK ---

FSIS began its first national baseline study in October 1992. The objectives of those early studies were fairly general, he said. FSIS wanted to collect enough data to provide a general microbiological profile of products, and to use that data and knowledge gained as a reference for further investigations and evaluations of new prevention programs. 

FSIS used the prevalence estimates for Salmonella on seven different products in developing the Salmonella performance standards. These baseline prevalences are estimates of the percentage of product that would test positive for Salmonella at the point in time when the baseline was conducted, he said. They can also be viewed as the probability that if you went out and took a sample of any of those commodities, you would detect Salmonella. 

Next, he said, FSIS decided that performance would be measured by a series of samples of a discrete set of size N. The Agency decided that an establishment operating at the exact baseline prevalence should have an 80% probability of passing one sample. The 80% was a judgment, Lange said, “a balance between the need to prevent the establishment from failing a set based on just pure chance and the need to identify establishments that are operating in a way which results in a Salmonella prevalence above the baseline.” 

Intervention Strategies
James S. Dickson, Professor and Chair of Microbiology at Iowa State University, chaired the panel that focused on the state of the scientific support for various intervention strategies, including verification of effectiveness. Providing a working definition of intervention as “a procedure, process, or technology which reduces or eliminates a potential foodborne hazard,” he reminded the audience that “The first intervention is prevention!”

Martin Firth, Manager for Policy and Strategies of the Canadian Food Inspection Agency (CFIA), discussed on-farm food safety interventions in Canada. CFIA, created in 1997 out of four separate government bodies, began work with on-farm commodity groups in 1995. In early discussions, producers expressed concerns that if they put something in place, they would be singled out, Firth said. However, through developing their programs, they soon realized that there’s actually more liability for doing nothing. He said that with the mandatory approach for HACCP in the meat and poultry sector, pressure is being put on livestock sectors to develop their programs. 

Animal production programs have to use sound science in their development, Firth said. CFIA realized very early on that, at the farm level, HACCP would be very difficult to implement in its pure sense or even as it is applied at the processing level. “So what we mean by HACCP-based is that these programs at a commodity level are developed through a technical committee,” he said. The technical committee must include some members from academia, technical experts, as well as producers and government personnel. It develops a generic model for the commodity and, using the seven basic principles of HACCP, goes through the hazard analysis based on that generic model. The committee identifies any critical control points (CCPs) and uses the hazard analysis for direction in the development of good production practices. The CCPs derived from this analysis are also incorporated into the good production processes. 

“The result,” Firth said, “is that the committee has not only a generic process that’s allowed them to develop their producer manual, but they also have a very clear record of their decision-making process.”

--- PAGE BREAK ---

John Sofos, Professor of Animal Science at Colorado State University, spoke on reduction of contamination at slaughter. He said, “One of the interventions is prevention, and by intervening before hide removal, we can reduce the initial contamination on carcasses that we will need to reduce by other interventions later in the process.” 

When there are heavily soiled or contaminated animals, he said, alternatives include segregating those animals and slaughtering them separately. This involves more plant employees and slows down the slaughter process. 

Chemical dehairing is effective because it keeps the hide contamination outside the plant. Sodium sulfide is used to hydrolyze the hair, and hydrogen peroxide to neutralize the sodium sulfide. The chemical dehairing process requires both a capital investment and a way to deal with chemical and hair waste. However, he said, the sodium sulfide can be regenerated and the hydrolyzed hair can be used to make fertilizer. 

Sofos said that a key concern with any intervention is whether it offers a potential for spreading or redistributing the bacteria if the intervention is not applied properly or if the equipment is not operating correctly. He cautioned, “We need to consider the potential long-term effects on interactions of different interventions on the microbial ecology of slaughter plants, raw products, and additive products. We need to find out why we have variation, and learn how to avoid it, and we need to remember that the product after slaughter is not ready to eat until it’s further processed or cooked.”

John B. Luchansky, Research Leader, Microbial Food Safety Research Unit at the Eastern Regional Research Center of USDA’s Agricultural Research Service, discussed characterization and control of foodborne pathogens. He said that ARS scientists ask many questions as a first step in developing research strategies for developing interventions. How does the pathogen respond to environmental cues such as salt, acid, heat, refrigeration, and what can you do to develop interventions with that knowledge? 

Predictive microbiology plays a big role. “The pathogen modeling program (PMP) is a group of models that estimate the behavior of pathogens in specific environments,” Luchansky said. “You can set the temperature, the pathogen, the salt concentration; and you can see whether or not that bacterium is likely to grow, merely survive or in some cases actually decline.” He said the PMP includes a reference database and allows users to go right from the PMP into the literature base. ARS has distributed several thousand CDs of the PMP, the program is downloaded from the ARS web site about 5,000 times a year, and ARS estimates that about 30% of the food industry use the PMP in designing HACCP systems. 

ARS began building a “companion” for the PMP about 18 months ago, Luchansky said. Combase, a collaborative effort with colleagues at the Institute for Food Research in Norwich, U.K., is a relational database of predictive micro information. “It is a very useful tool for academicians and risk assessors because it actually contains the raw data. You can get right in and access raw data on which models might not yet be available,” he said. 

A key question too often overlooked in developing interventions is “Where did the pathogen come from, and where might it end up?” Luchansky said. “A perfect is example is livestock. Are the pathogens indigenous to the livestock, or did they come in on the feed? Were they dropped on the farm by birds and then consumed by livestock? We really do need a lot of on-farm microbial ecology and in-plant microbial ecology.” 

He discussed several studies ARS has conducted or collaborated on whose results are beginning to answer that question. In one study, ARS found a similar prevalence of E. coli O157 isolates in swine colon samples and fecal samples, yet recovered serotype E. coli O157:H7 isolates from the colon but not the feces. “This finding raises new questions about the impact of collection, storage, shipment, and/or methodology,” he said. “It raises questions about the impact of transport and holding on the shedding and/or viability of E. coli O157:H7. It provokes the question, Should further studies be initiated to determine the prevalence of the pathogen in matched animal and fecal samples from the farm all the way to slaughter?”

--- PAGE BREAK ---

Michael Ollinger, an economist with USDA’s Economic Research Service, spoke on the costs and benefits of adopting food safety interventions. Quite simply, he said, firms are going to invest in food safety technologies only up to the point where it’s profitable. As soon it becomes unprofitable, they’re not going to invest. “If quality sends a clear signal to the consumer or to the ultimate buyer, then the market is going to work,” he said. 

One market approach to controlling food safety is the unintended-consequence investment, Ollinger said. For example, a poultry plant might invest in some kind of poultry transfer mechanism to reduce the amount of labor on the production lines. They may also receive a “savings” in bacterial contamination because employees handle the poultry less. A seller can recover all or most of the profit from a food safety intervention, such as irradiation, if he or she can communicate directly to the consumer that food is safer because of the intervention. Producers of branded products have a lot to lose by providing an unsafe product, so they’re perhaps going to invest a little more in food safety to ensure they don’t lose their market if someone gets sick, he said. The major fast-food restaurants and the major restaurants now have explicit buyer contracts, another type of successful market approach. Suppliers are willing to expend some costs to meet higher standards if it will increase their sales volume. 

However, Ollinger noted, there are some problems with market approaches. For example, a retailer may receive meat products from more than one provider. If the product is recalled, it may be impossible to determine the source of the product that made a person sick. If the market doesn’t work, he concluded, then government intervention may be called for. 

Conferences Further Food Safety
Merle Pierson, USDA’s Deputy Under Secretary for Food Safety, closed the conference by saying that dialogue such as this important discussion of the science is needed in establishing food safety policy. There has been much progress over the years with the advent of new surveillance methodology, microbial identification techniques, processing technologies and intervention strategies, and approaches to food safety management. The recent report by CDC on foodborne illness provides encouraging information indicating significant declines in foodborne illness and the success of food safety interventions over the past several years. “We have also seen rapid advances in the area of risk analysis in the support of policy decisions,” he said. 

“It is clear that adoption of HACCP as a food safety management system has been important to improving the safety of our food supply,” Pierson said. “However, HACCP is only a food safety management tool that must be supported by essential scientific information.” Although there is much scientific information available to support the development of HACCP systems and to establish food safety policy, he said, there are still many critical scientific questions that need to be answered. 

“We need to have a clear understanding of the relationship of food safety policy and hazard management to public health outcomes,” he said. “For example, what is the relationship of specific levels and incidence of pathogens such as Salmonella in raw meats to foodborne illness? To what level can these pathogens be reduced in raw meat and poultry, and what is the associated impact on public health? Are there certain serotypes/biotypes, and associated ecological niches from primary production to consumption that have the most significant contribution to foodborne illness? Once these niches are determined, what are the most effective intervention points and strategies for pathogen reduction and associated impact on public health?” 

“How should performance standards be used, and how do they relate to public health outcomes?” Pierson asked. “What are the handling practices that have the greatest impact on food safety and how can they be improved?” 

“It is clear that we need multiple approaches to be effective in reducing foodborne threats to public health,” he said. “However, these approaches must be appropriately targeted and the true impact clearly understood.” He concluded by saying that “Food safety is the responsibility of every person involved in the food system, from primary production to final use of the food. It is through conferences such as this one that we can work together and exchange essential information that is needed for furthering food safety.”

--- PAGE BREAK ---

Agriculture Secretary keynotes conference
Secretary of Agriculture Ann M. Veneman keynoted the conference by saying, “We have made significant strides in food safety in recent years. The implementation of the Pathogen Reduction/HACCP rule stands out as a major achievement, and it has played a vital role in reducing pathogens. Data released just last month by USDA show that Salmonella prevalence levels are well below the product baselines that existed prior to HACCP. This decrease correlates with reports from the Centers for Disease Control and Prevention indicating a decline in Salmonella-related human illnesses during the same time period—tangible evidence of our success.

“We must remember that this achievement has behind it decades of hard work and dedication from individuals who have spent lifetimes in the pursuit of making food safer to eat,” she said. She presented the first Howard E. Bauman Award for Food Safety to Bauman’s daughters Victoria Zobel and Kay Rose on behalf of their father, who died in August 2001 (see Food Technology, June 2002, p. 126, for more information on the award). Bauman, who was IFT President in 1977–78, played a significant role in the development of the Hazard Analysis and Critical Control Point System (HACCP). The award was created to recognize individuals showing longstanding commitment, leadership, and service to improving food safety. Veneman noted, “We owe much to Dr. Bauman for his creativity, and I’m sure he would be pleased to know that HACCP is now in place in meat and poultry plants nationwide. It is because of his vision and foresight that American consumers today have a higher level of food safety protection.”

“Hard work and determination by numerous scientists, government officials, academicians, consumers, and the private sector, contributed to building this system,” she said, “but we cannot rest on our laurels—much remains to be done. Science has served as the foundation for these successes, and we are here today to continue this progress.”

Sessions Provide Lessons
Speaking to the panelists and audience the morning of the second day of the symposium, Elsa A. Murano, Under Secretary for Food Safety, noted: “Why is FSIS engaging in these scientific symposia? How are these meetings different from others the Agency has held in the past? . . . Policy makers need to make the best decisions possible. Decisions that will address the underlying problems affecting food safety. Decisions that will provide solutions which can be measured in terms of public health. In my opinion, these decisions must be based on science, and not on the path of least resistance.”

Murano said, “There are three lessons I came away with from yesterday’s sessions. One is that prevalence data derived from regulatory testing is by its very nature biased data, and will not provide us with the true incidence of pathogens in meat and poultry. Second, data on foodborne illnesses collected by CDC are incomplete and do not provide adequate information on the contribution of various factors—such as the type of food involved—on disease. Both of these are essential if we are to determine whether interventions, HACCP, or other factors are making an impact on public health. Third, yesterday’s meeting helped put into perspective which of the steps from farm to table are the key points where contamination must be controlled in order to improve food safety.”

by Sharin Sachs and Karen Hulebak
The authors are, respectively, Policy Analyst, Executive Management Staff; and Deputy Administrator, Office of Public Health and Science; Food Safety and Inspection Service, U.S. Dept. of Agriculture, Washington, DC 20250. Send reprint requests to author Sachs.