Statement of

Ann Hollingsworth, Ph.D.

President, American Meat Science Association

On behalf of

American Meat Institute and

American Meat Science Association

before the

Senate Agriculture Committee

September 20, 2000

Thank you for inviting me to testify on the best ways to reduce microbial contamination in the U.S. food supply. As a meat scientist and a meat industry executive, I have worked in this arena for more than 20 years. I am speaking today on behalf of the American Meat Science Association, a professional society of 1000 meat scientists and the American Meat Institute, the nation's oldest and largest trade association representing beef, pork, lamb, veal and turkey slaughterers and processors. Based on my training and experience, I'd like to make a few observations at the outset, followed by three key messages.

First, let me share my observations about meat's microbiological safety in the year 2000:

· Pathogens have truly become "public enemy number one" for everyone in the meat processing industry. Twenty years ago the enemy was fat; forty years ago the enemy was animal diseases. But today, it is definitely pathogens. Based on our success fighting animal diseases and too much fat, I expect we will win our "war on pathogens" and move onto some other battlefront before too long.

· Meat has less bacteria of all kinds - harmful and benign - today than it did even 10 years ago. Government and industry surveys show reductions in Salmonella and generic E. coli on raw meat and poultry, and Listeria on ready-to-eat meat and poultry, over the past 10 years.

· Meat slaughtering and processing companies have better food safety technologies and training in place today than ever before, and these have made the major difference in food safety improvements.

· Foodborne illness rates are decreasing, safe food handling awareness is increasing, both of which contribute to better public health.

From my perspective, I definitely see the "glass half full" when it comes to the microbiological safety of meat.

I have three messages to leave with you today:

· First, on the much-discussed topic of microbiological testing, I must tell you that it is an important tool for verifying good process control in a plant, but on its own it does nothing to assure food safety. So it is part of something larger -- and should never be viewed as a "stand-alone" segment of food safety systems.

· Second, we should all be looking harder at technology and education as the true keys to reduced microbial contamination in our food supply. Both the public and private sectors could do more practical research and development, at every segment of the food chain, to identify technological and educational solutions to contamination problems.

· Third, government has not always managed its regulatory resources well in the war on pathogens. For example, declaring pathogens illegal in raw agricultural products (such as E. coli 0157:H7 in ground beef) has discouraged industry testing, given consumers a false sense of security and given some in government and industry the false "crutch" of microbial testing to lean upon instead of developing better control measures. Government should reevaluate its efforts to reduce microbial contamination -- in concert with the scientific, public health and industry communities.

Microbiological Testing - An Important Food Safety Tool

Microbiological testing has been used for decades in the food industry - including the meat industry - to verify good plant hygiene or good process control. In fact, 32 meat scientists from government, industry and academia developed a consensus paper in January 1999 on "The Role of Microbiological Testing in Beef Food Safety Programs." The paper is available through the American Meat Science Association and I am submitting a copy with my written testimony. My comments on microbiological testing are largely derived from that paper.

Most scientists agree that successful microbiological testing programs must be associated with achievable and verifiable microbiological criteria. So microbiological criteria - such as the absence or presence, or limited presence, of a microbe - are necessary. And to underscore what may not seem as important to you but are actually critical: the criteria must be achievable and verifiable. In other words, if it's not possible to achieve or measure, it won't work.

Interestingly, the National Research Council assembled a panel of experts to develop recommendations on microbiological criteria for foods in the early 1980s. Their 1985 report addressed 22 groups of foods and food ingredients -- and for raw meats, the experts did not recommend establishing microbiological criteria because such criteria would neither prevent spoilage nor foodborne illness. According to the panel, pathogens of public health concern are often present in small numbers as part of the natural microflora of live animals, and in 1985 - as today - could not be totally eliminated through animal husbandry and meat processing techniques. Therefore, the panel stated that it would be impractical to set limits for microbiological pathogens in raw meats as it would be impossible to comply consistently with the limits.

Instead, the NRC panel recommended 1) a recognition that low levels of pathogens may be present on raw meats; 2) strict adherence to good food preparation practices; 3) application of new processing procedures designed to reduce the presence of pathogens; 4) education on food handling practices; and 5) implementation of HACCP.

Both AMI and AMSA agree that microbiological sampling is a useful food safety verification tool in the context of a total food safety process control system. The groups also agree that microbiological criteria are useful in a food safety system as long as they are both achievable and verifiable. However, the groups do not support punitive action from USDA if a plant fails to meet microbiological criteria that do not measure plant hygiene or product adulteration.

Technology and Education - Keys to Reduced Microbial Contamination

Just in the past 10 years, the meat industry has made phenomenal reductions in microbial contamination through the use of new technologies. In the beef and poultry sectors, carcass washing and rinsing technologies involving various combinations of heat, cold and chemicals have reduced microbes by up to 95 percent or more. I know it may be hard to believe, with all the publicity over E. coli-related beef recalls, but the products today are actually more microbiologically safe than they were 10 years ago.

Education and heightened awareness plays a tremendous role in all manufacturing industries. Listeria control is a perfect example. We don't have any significant new technologies for Listeria control - yet - but through vigilant sanitation efforts, the incidence of this pathogen in both hot dogs and lunch meats has dropped dramatically over the past 10 years. Again, headlines in the press would lead you to a different conclusion, but according to USDA data, the incidence of Listeria in lunch meats dropped from 10.7 percent to 4.6 percent between 1989 and 1999; and the incidence in hot dogs dropped from 7.9 percent to 1.8 percent in the same time period.

And let's not forget consumers and professional food preparers. They have received intensive education about safe food handling from both the public and private sectors over the past 10 years. Those efforts are paying off, with surveys showing consumers more aware than ever of safe food handling and preparation methods.

Some Suggestions for Government Food Safety Improvements

I believe the federal government shares the concern we all have about reducing microbial contamination of foods. I also believe the government has tried to make a positive difference in the safety of the food supply. But, from my perspective, the government in total and USDA's Food Safety and Inspection Service in particular have made some wrong turns and sent some mixed signals that, in some cases, have probably inhibited food safety improvements.

I would like to close by offering seven suggestions for consideration by FSIS that I believe would lead to more constructive use of microbiological criteria and a reduction in microbial contamination in meats.

The challenge for FSIS is how to incorporate a microbiological monitoring and surveillance system into an equitable regulatory scheme that is designed to reduce the pathogen level on raw meat and poultry products. Some points need to be made in this regard:

1. A pathogen monitoring and surveillance system must be designed to measure trends over time. Statistical process control techniques or other appropriate statistical analysis should be used to evaluate the data. For example, an upper limit could be set at three standard deviations above the mean to accommodate normal process variability. Two consecutive data points that exceed the upper limit could trigger an investigation of the plant's control programs. (USDA currently measures performance over time in its existing Salmonella performance standards, however, the principle is not based on the principles of statistical process control. Data feedback is too slow and unresponsive for a plant to effect meaningful change.)

2. In theory, each plant should set its own microbiological criteria to accommodate normal process variability, but that would create unjustified inequities between plants regardless of the degree of control exercised by the plant. Therefore, microbiological criteria should be established to allow for differences that are primarily attributable to the live animals entering the slaughter facility. For example, USDA could establish criteria that are based on seasonal, regional, species and class differences. (USDA accounts for species and class differences but not for regional and seasonal differences in its Salmonella performance standards.)

3. If microbiological criteria are established to account for seasonal, regional, species and class differences, it must be recognized that the primary objective is to reduce the overall incidence of pathogens in the raw meat and poultry supply. Except for E. coli O157:H7 in ground beef, the mere presence of pathogens on raw meat and poultry products does not legally render the product to be adulterated nor does it make the product unfit for human consumption. Therefore, the appropriate regulatory response for a plant that repeatedly does not meet the pathogen performance standard is to require the plant to reassess its control programs and implement corrective action, as needed. (USDA requires a plant to reassess its control programs and implement corrective actions if the plant fails to meet the codified Salmonella performance standard; but three consecutive failures will result in suspension of USDA inspection. The suspension of inspection based on a failure to meet the Salmonella standard is currently being litigated.)

4. USDA should also conduct an audit of a plant that repeatedly fails to meet the microbiological criteria to determine if the plant is manufacturing product in a sanitary manner. A finding by USDA that the plant is producing product under insanitary conditions would cause the product to be adulterated and inspection services suspended. (USDA has the statutory authority to suspend inspection if product is produced under insanitary conditions, but the court has ruled USDA cannot suspend inspection based on a failure to meet the Salmonella performance standard, which does not measure plant sanitation.)

5. Pathogens that are present in ground product usually originate from the raw materials used to produce the ground product. The only practical means grinding operations have for reducing pathogens is to control the source of raw material they purchase. Proper sanitation and temperature control can retard pathogen growth, but grinding operations cannot reduce pathogens that are present in the raw materials. Therefore, a federally mandated microbiological monitoring and surveillance program should concentrate its resources on slaughter operations to reduce pathogens as far back in the supply chain as possible. (USDA's Salmonella performance standard requires both carcass and ground product sampling, but emphasis is placed on ground product sampling in plants that produce both products.)

6. The rate of pathogen testing should be based on the number of carcasses processed in the facility. Samples should be collected throughout the year to properly assess seasonal differences. For example, very small plants that slaughter only a few animals per year could be tested once per quarter and large plants could be tested weekly. The objective is to sample at a rate that is roughly proportional to the amount of product a plant produces. (USDA's current sampling rate is based on completion of data sets without regard to the samples being randomly distributed throughout the year.)

7. In the final analysis, the fundamental question is whether USDA should have the authority to suspend inspection and prevent a plant from operating if the plant fails to meet a pathogen performance standard. That is a public policy question that relates to the inability of a plant to meet a microbiological criterion. It is not a scientific question, but conventional wisdom would dictate that if a determination cannot be made that the product is adulterated and unfit for human consumption, then the plant should be allowed to operate with additional government scrutiny and oversight.

Thank you again for this opportunity to testify. I would by happy to answer any questions.