Remarks prepared for delivery by Dr. H. Scott Hurd, USDA Deputy Under Secretary for Food Safety, to the Iowa State University, 5th Annual BIGMAP Symposium, panel on Biotechnology Innovation and Regulatory Changes, on April 23, 2008, in Des Moines, Iowa.

Notes: Slides are available in an attached PDF document; individual pages are linked within the text. Links to organizations outside USDA open in a second window.

Thank you. I appreciate the opportunity to participate in today's program and especially in this afternoon's panel. It's nice to be back at Iowa State and to see some familiar faces.

I'm the Deputy Under Secretary for USDA's Office of Food Safety. We oversee the Food Safety and Inspection Service, which carries out USDA's food safety regulatory program, as well as important public health outreach and education activities focused on enhancing the safety of the U.S. food supply.

(Slide 2) Our mission is to ensure the safety and wholesomeness of the nation's commercial supply of meat, poultry and processed egg products. It doesn't matter if those products are imported to, or exported from, the United States.

We're also dedicated to fostering safer food handling habits among all types of food handlers, and developing science-based policies to improve our food safety and defense systems.

Acts and Authorizing Statutes
(Slide 3) Our vital food safety activities are conducted under a number of federal acts and authorizing statutes. The three that most clearly define our role in food safety are the Federal Meat Inspection Act of 1906, the Poultry Products Inspection Act of 1957, and the Egg Products Inspection Act of 1970. They provide the basis for FSIS' ability to perform as a public health agency.

To ensure that the foods we regulate are safe, FSIS employs more than 9,000 employees, including approximately 7,800 inspection personnel who are in more than 6,000 establishments every day/every shift.

(Slide 4) FSIS' sister agency, the Department of Health and Human Services' Food and Drug Administration (FDA), is responsible for all other food products not under FSIS' jurisdiction. This can include fresh produce, ice cream, and seafood.

I believe this system works well and that the nation's public health is well protected. The combined efforts and expertise of the different federal and state agencies allow quicker resolution to outbreaks and crises than any one of the agencies could accomplish working by itself.

Public Health Focus
USDA believes that to have the greatest impact in preventing foodborne illness, we must focus on the products and processes that have the most risk.

FSIS has incorporated risk into a number of agency initiatives and policies and we've taken a risk-based approach to battling pathogens that cause foodborne illness. (Slide 5)

• Risk analysis plays a vital role in the USDA/FSIS food safety program.
  • Risk assessments provide a scientific basis for food safety policies and allocation of inspection resources.
  • FSIS corporate strategic plan/SOPs.
• Powerful public health tool
  • Allows for transparency and stakeholder involvement to ensure credibility and scientific accountability.

(Slide 6) When a proposed regulation is expected to have a significant impact on the regulated industry, FSIS is required to conduct risk assessments, under the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS Agreement), World Trade Organization (WTO) guidelines, and the Office of Management and Budget's (OMB) Executive Order 12866, along with a cost-benefit analysis. Executive Order 12866 requires a risk assessment for any regulation expected to impact industry by $1 million or more annually.

Most FSIS risk assessments developed, to date, have been as part of the regulatory process (required under Executive Order 12866). These risk assessments require substantial data input and critical analysis to meeting the regulatory requirements of the Information Quality Act (IQA).

As part of the IQA requirements, these risk assessments are objectively peer reviewed by scientists external to the agency. The peer review comments and agency response are posted on the FSIS Web site and made readily available to the public.

An additional aspect of the Agency's effort to be transparent is in the solicitation of public comment, often through public meetings and posting documents on the Web site. The Agency values input from stakeholders on both its risk assessment agenda and the ultimate risk assessments produced. The Agency actively seeks additional information and data to refine its risk assessments continually to better inform its decisions.

(Slide 7) FSIS develops food safety risk assessments to inform various agency decisions including:

• the development of regulations,
• agency recalls,
• food security vulnerabilities, and
• science-based equivalence criteria.

Risk assessments provide a scientific basis for the agency's food safety decisions and enable us to explicitly link policies to public health outcomes.

(Slide 8) USDA/FSIS has been developing quantitative microbial food safety risk assessments since the mid-1990s. Some of the first quantitative microbial risk assessment models for food safety were those developed by USDA/FSIS in the mid-1990s:

• Salmonella Enteritidis in shell eggs and egg products (done in collaboration with FDA),
• the BSE risk assessments and, perhaps the most comprehensive model,
• a farm-to-table E. coli risk assessment that was not only comprehensive in scope (an exposure assessment modeling the growth and decline of E. coli O157:H7 from farm to table based on changes in environmental conditions (pH, water activity, time, temperature, etc.), but also because this model included 2nd order modeling to separate uncertainty and variability in the model. Because of the complexity in conducting a 2nd order model for uncertainty analysis, most risk assessments conduct a sensitivity analysis in place of 2nd order modeling to evaluate uncertainty in a microbial risk assessment model.

(Slides 9-10) FSIS has completed several QMRAs used to guide the development of food safety standards, consumer education efforts, and the allocation of inspection resources.

(Slide 11) As an example, I'd like to talk about our risk assessments related to Listeria monocytogenes and the role they played in the agency's move to risk-based sampling. These risk assessments ultimately laid the foundation for an Agency Listeria Action plan, much of which is in implementation today.

Why focus on Listeria monocytogenes? (Slide 12)

• >90% hospitalization rate, >20% fatality rate; occurs most often in immune compromised individuals (cancer patients, transplant recipients).

Lm is often fatal to elderly and causes miscarriage in pregnant women. This bacterium is killed by typical cooking/pasteurization temperatures. Lm is nearly ubiquitous in the environment (both agricultural and processing), and therefore is a recontamination concern for RTE foods. Unlike most other foodborne pathogens, Lm grows at refrigeration temperatures.

Lm is a concern in RTE foods which do not receive a heat treatment immediately before consumption, especially processed foods with high moisture and moderate pH (like soft cheeses, seafood, and deli meat).

Data gathered after outbreaks of Listeria-related illnesses during 2000, 2002, along with the information published by the Centers for Disease Control and Prevention (CDC), combined with food safety investigations and in-depth verification reviews, led FSIS to conclude that some establishments were not adequately addressing the potential for bacterial contamination in their Hazard Analysis and Critical Control Point (HACCP) plans, Sanitation Standard Operating Procedures (SSOPs), or other control measures.

This information prompted FSIS to initiate the first of three formal Risk Assessments to illuminate the public health problem of listeriosis and identify appropriate risk management steps. (Slide 13)

These are the three overarching risk management questions related to Listeria and the associated risk assessments the Agency completed to answer them.

• Which ready-to-eat foods pose the greatest risk of listeriosis?
  • 2003 FDA-FSIS Quantitative Assessment of the Relative Risk to Public Health from Foodborne Listeria monocytogenes Among Selected Categories of Ready-to-Eat (RTE) Foods
• Which interventions effectively control Lm?
  • 2003 FSIS Risk Assessment for Listeria monocytogenes in Deli Meat
• How can FSIS target its inspection resources to effectively address Lm?
  • 2005 FSIS Risk-based Verification Sampling for Lm in ready-to-eat (RTE) meat and poultry products

(Slide 14) After the first assessment identified RTE meat and poultry products as primary contributors to the listeriosis problem in the United States, the agency focused on "Which processes control effectively for Lm?" or "How can FSIS effectively reduce that risk for the consumer through the regulation of meat and poultry production?"

The 2003 FSIS Listeria Risk Assessment defined risk factors for Lm contamination and the related processes that would produce these RTE products in safer ways:

• post-packaging lethality (like pasteurization),
• antimicrobial agents incorporated into the formulation and antimicrobial processes (such as fermentation),
• specialized sanitation, and
• surface and final product testing.

In 2003, FSIS released this scientific risk assessment on Listeria in RTE meat and poultry products. A public meeting was held on February 26, 2003, to discuss the risk assessment. The risk assessment, in conjunction with a previously released FDA/FSIS risk ranking and public comment gathered on the topic and a peer review of the risk assessment, provided important data enabling FSIS to design a final Lm rule.

(Slide 15) The FSIS Listeria risk assessment was designed to address three specific risk management questions:

• What is the effectiveness of testing and sanitation of food contact surfaces on mitigating product contamination and reducing the subsequent risk of illness?
• How effective are other pre- and post-packaging interventions in mitigating product contamination and reducing the subsequent risk of illness?
• What guidance can be provided on testing and sanitation of food contact surfaces for Listeria species?

To address these risk management questions, a dynamic in-plant model (referred to the in-plant model) (Slide 16) quantitatively characterizing the relationship between Listeria species in in-plant environment and L. monocytogenes in deli meats at retail was developed using the available data. The outputs of the in-plant model (e.g., concentration of L. monocytogenes on deli meat at retail) were used as inputs into the updated FDA/FSIS retail-to-table exposure pathway for deli meats. These two connected models — the in-plant model and the updated retail-to-table FDA-FSIS exposure assessment and FDA/FSIS dose-response relationship — comprise the overall risk assessment model.

(Slide 17) Combinations of interventions (e.g., testing and sanitation of food contact surfaces, pre-and post-packaging interventions, and the use of growth inhibitors/product reformulation) appear to be much more effective than any single intervention in mitigating the potential contamination of RTE product with L. monocytogenes and reducing the subsequent risk of illness or death.

(Slide 18) Interim Final Rule (issued June 2003, effective October 6, 2003; 9 CFR 430 RTE)

• Based on the 2003 FSIS Listeria risk assessment and the 2003 FDA-FSIS Risk-Ranking for Lm in ready-to-eat foods.
• Established three alternative control measures:
  • Post-lethality treatment and antimicrobial agent/process with sanitation (Alt. 1);
  • Post-lethality treatment (Alternative 2a) or antimicrobial agent/process with sanitation (Alt. 2b);
  • Sanitation alone (Alt. 3).
• FSIS Directive 10,240.4 (10/2/03)
  • Established a risk-based sampling program to verify compliance with Lm Rule.

(Slide 19) The risk-based sampling program was designed:

• To prevent adulteration of product by Lm.
• To focus USDA/FSIS verification sampling efforts on establishments producing product predicted to pose the greatest risk of listeriosis based on two previously developed Listeria risk assessments and real time laboratory data.
• To evaluate an establishment's food safety system so that FSIS can focus on establishments with less effective Lm control measures.
• To provide an appropriate incentive for establishments to adopt more effective Lm control measures.
• To protect public health.

(Slide 20) Risk-based verification sampling was implemented in January 2005. Previously, establishments were tested randomly; now, using the risk ranking algorithm, frequency of sampling is dependent on relative risk for Lm contamination.

• Approximately 10,000 product samples annually.
• Expanded to collect product, environmental and food contact samples in April 2006.
• Risk is defined quantitatively for each facility.

(Slide 21) Risk-based initiatives work as shown by this graph. Since 1998, the percentage of regulatory samples of meat and poultry products that tested positive for Lm has fallen by almost 80 percent.

(Slide 22) We also used a risk-based approach to reduce the incidence of Salmonella on raw products. Since announcing a targeted 11-point Salmonella reduction plan in February 2006, FSIS has seen consistent declines in product positive rates.

Since July 2006, FSIS has used a risk-based algorithm to determine priority for Salmonella set scheduling at eligible establishments. Of the 75 sets scheduled every month, new plants (including new plant/product combinations) are the first priority, followed by Category 3 plants (i.e. those that exceeded the performance standard in their last set), then Category 2 plants, and Category 1 plants.

In broilers we've seen the numbers go from more than 16 percent of samples testing positive for the pathogen in 2005 down to single digits. Building risk-based efforts on science works.

(Slide 23) This chart shows the impact this initiative has had on establishment categories.

• Category 1: 35% to 75%
• Category 3: 10% to 3.5% (7 plants)

On March 28 we began publicly listing plants by category, making it clear to everyone — the news media, business partners, and ordinary consumers — which ones are poorly performing. We've started with young broilers, but are considering other product categories as well.

Over the long haul, we've made some tremendous progress in controlling E. coli O157:H7. Between 2002 and 2006, FSIS testing shows the percentage of samples testing positive for E. coli declined by 80 percent. Maybe not the best indicator of prevalence, but certainly an indicator of trends and increasing industry control of this bug — until recently.

(Slide 24) This slide is slightly different than the ones I showed you for Listeria and Salmonella. The Y axis on the left shows the percent positive regulatory beef samples for E. coli O157:H7, indicated by the blue bars. The Y axis on the left shows the number of E. coli O157:H7 illnesses reported to the CDC, indicated by the red diamonds. The changes in the number of percent positives from 2003 to 2007 are not statistically significant. However, you will notice that the trend for percent positives is similar to that for reported illnesses.

The agency's E. coli O157:H7 initiatives and industry's collective response in 2002 helped drive the rates of positive samples down in 2002, 2003 and 2004 — and these rates remained at 0.17 percent for 2005 and 2006. But in 2007, the rate increased to 0.23 percent. To put that percentage into perspective, out of 12,000 samples taken in 2007, only 27 — a miniscule amount — were positive for E. coli O157:H7.

Where Should We Be Headed?
(Slide 25) We are committed to making the foods we regulate the safest they can be, wherever they are produced, large or very small, urban or rural, domestic or abroad. We're continually exploring ways to more efficiently perform this vital inspection role and apply our resources where we can be most effective and have the most impact on protecting public health.

To be successful, these public health decisions must be based on information. The agency has made great strides in collecting data, analyzing it and responding to it — and is working to use data to predict problems better before they occur to further strengthen the public health protection.

Continued data infrastructure development will make data collection, analysis, and reporting easier and quicker at all levels within FSIS, and provide the foundation upon which the agency's public health decisions are made.

We are currently developing a Public Health Risk-Based Inspection System. The goal of this science-based and data-driven system is to focus our resources where they can best ensure food safety systems are under control. That focus, we believe, will help us achieve FSIS' public health mission.

We'll have inspectors spend most of their time where they can do the most good — at establishments making high risk products and/or exhibiting poor controls of risk.

Another initiative focuses on completing Food Safety Assessments (FSAs) in all plants, using an objective scoring mechanism to help determine the level of inspection needed. FSAs are conducted by specially trained Enforcement Investigation and Analysis Officers and give us our best in-depth analysis of an establishment's risk control measures.

(Slide 26) Higher risk products and processes demand a continuous inspection presence — with the federal government on the premises — to ensure measures are in place and put into action to control pathogens, lowering the likelihood that you or your family get a foodborne illness. Compare raw ground poultry to chicken soup. They get the same amount of inspection now, but we're changing that. Risk determines (or should determine) inspection intensity and frequency.

Closing
Ensuring the safety of meat and poultry products, given the realities of today's modern supply chain, requires that we have the ability to anticipate, and more quickly respond to, food safety challenges before they negatively affect public health.

Events such as this one enable us to focus our combined energies on solutions to the challenges we face. I'd like to thank Iowa State University again for holding today's meeting and for inviting me to participate.

USDA will continue to engage the scientific community, public health experts, food safety partners, Congress and all interested parties in our efforts to identify science-based solutions to public health issues to ensure positive public health outcomes. We all know that we can save lives with sensible science-based policies and together we'll do just that.

Now I'd like to take your important questions.