1. The FSIS Role in Food Safety
Who’s Minding the Store? The Current State of Food Safety and How It Can Be Improved
David Goldman, MD, MPH
Assistant Administrator
Office of Public Health Science
Food Safety and Inspection Service
April 11, 2008

2. FSIS Mission
Protect consumers by ensuring that meat, poultry, and egg products are safe, wholesome, and accurately labeled.
With FDA, responsible for ensuring the Healthy People 2010 food safety goals
Incorporate food safety objectives into agency strategic plan and OMB reporting

3. FSIS Workforce
10,000 total personnel—of that number 7,500 are inspection and veterinary personnel
6,300 plants have FSIS inspectors present in them every day, as required by our acts
100 billion pounds of meat, poultry and liquid egg products are verified safe by FSIS each yr
Inspects product that represents more than one-third of all consumer spending on food in the U.S. and about 40% of all domestic food production

4. FSIS Profile
Forerunner organization established in 1906 with passage of the Federal Meat Inspection Act
Agency has historically been focused on inspection activities
Public health mission has evolved more recently

5. FSIS Profile HACCP Rule passed in 1996
Focus on verification of establishment pathogen control
Increased product testing
Collaboration with PH partners in foodborne illness investigations
Founding partner of FoodNet in 1996
Member of PulseNet

6. FSIS Application of Public Health Model
Quantitative risk assessments
Other scientific assessments
Food data
Human data
Animal data
Environmental data

7. HP2010 Food Safety Goals
Reduce infections caused by key foodborne pathogens; targets were set at 50% of the illness rate in 1997

8. FSIS Initiatives for E. coli O157:H7
FSIS target = 0.2% prevalence in ground beef ( rates each 0.17%; preliminary 2007 rate=0.24%)
Recent baseline in beef trim (major component of GB)
Routine trim testing
More aggressive follow-up testing of positive tests
Risk-based sampling based on testing history and volume
New micro testing methodology
Update to E. coli O157 risk assessment:
model both the contribution of all ground beef components; and,
effect of industry interventions
Consideration of primal cuts and non-O157 STEC

9. FSIS Initiatives for Salmonella
February 2006 announced new sampling/testing strategy
Focus initially on broilers, having increasing contamination rates and known vehicle for human illness
More agency testing would be done in plants with higher rates of contamination or harboring serotypes of human health concern
Agency goal: move plants to lower contamination rates
This new approach resulted in a change in our performance measure as reported to OMB
From: overall percent of samples positive for Salmonella
To: percent of plants performing at desired level

10. FSIS Initiatives for Listeria monocytogenes
FSIS target: 0.70% contamination rate (preliminary 2007 rate=0.37%)
Risk assessments revealed that deli meat and hot dogs were greatest risk among FSIS-regulated products
Risk assessments also showed the benefits of industry mitigations, adding growth inhibitors, antimicrobials or other post-lethality treatment to ready-to-eat products that may be exposed to Listeria in the environment
Pathogen testing became risk-based, factoring past testing, volume of production, use of interventions, and product risk
Recent study suggests that cross-contamination in retail delis may be a significant source of Lm contamination

11. FSIS Initiatives for Campylobacter
Lab methodological problems delayed our ability to establish a performance standard or begin a regulatory program
Baseline testing in broiler chickens currently underway—Campy historically associated with chicken exposure
Once complete the agency will establish performance standards or guidelines for industry to meet
Industry interventions for Salmonella expected to be effective against Campylobacter

12. Other FSIS Initiatives
Hosted attribution public meeting in April 2007
Hosted public meeting on non-O157 STEC in October 2007
Hosted public meeting April 9-10 on STECs
Will host public meeting May 15 on improving communications/collaborations during outbreak investigations
New investigations directive to be released, regarding how FSIS investigates foodborne illness linked to regulated products, the type of evidence collected and how it is judged
Strengthened PulseNet and VetNet database linkage
Development of public health-based performance measures

13. Re-assessment—Case Study #1
Outbreak of E. coli O157:H7 associated with frozen steaks in 2003
7 cases of illness
Steaks sold door-to-door
Traceback to producer revealed these steaks were tenderized/marinated
Food Safety Assessment conducted; injection equipment disassembled weekly
Recall of one week’s production of beef products
This case resulted in enforcement (NRs plus recall), regulatory policy, and a public health outcome.
Regulatory action: enforcement; policy development
Public health action: policy; increased knowledge of a PH problem

14. E. Coli case, continued FSIS Regulatory Policy Response
Federal Register Notice published 5/26/05
Cited 3 outbreaks in 2000, 2003, 2004 and total of 17 cases associated with exposure to tenderized beef products
The FR Notice announced that:
Plants producing such products should reassess their HACCP plans during their annual reassessments to consider whether E. coli O157:H7 is a hazard reasonably likely to occur
Plants should consider purchase specifications for source products treated to minimize E. coli O157:H7
Plants should consider use of antimicrobial treatments prior to tenderizing

15. E. Coli case, continued
Collaboration with FDA to develop guidance for retailers for incorporation into the Food Code
Development of FSIS compliance guidelines
Consideration of labeling changes and need for higher temperature for cooking
FSIS inspection personnel recently surveyed the industry to determine the extent and nature of HACCP reassessments
FSIS has three field service laboratories, in Athens, GA; St. Louis, MO; and Alameda, CA.
Field service laboratories help protect the health and welfare of consumers by ensuring, through analysis of samples collected at regulated facilities, that meat, poultry, and egg products are safe, secure, and not adulterated/misbranded.
They conduct baseline and other studies to identify the risks/vulnerabilities associated with meat, poultry and egg products from food borne hazards and intentional acts of terrorism.
The labs provide data to evaluate and verify FSIS’ effectiveness and efficiency in achieving its public health responsibilities.
Ultimately the work of the laboratory affects the health and safety of consumers worldwide.
All of the FSIS field service laboratories are accredited under ISO

16. E. Coli case, continued
Request for FSIS Office of Public Health Science (OPHS) to consider collecting data on prevalence of this pathogen in these products (in collaboration with Agricultural Research Service)
Request for OPHS to update its risk assessment on E. coli O157:H7 in non-intact beef products
There were 3 more outbreaks related to non-intact beef products in 2007

17. Re-assessment—Case Study #2
Salmonellosis associated with frozen stuffed raw poultry products
Illnesses in 2005 linked to breaded, frozen chicken entrees
FSIS issues a public health alert reminding consumers that these products contain raw poultry, though the breading has been browned, that microwave cooking can be inconsistent, and these products should be cooked to 165o F internal temperature
FSIS asked the National Advisory Committee on Microbiological Criteria for Foods (NACMCF) to recommend a safe cooking temperature for poultry

18. Salmonellosis case, continued
More cases of salmonellosis linked to similar products reported in early 2006
Investigations showed that some of the illnesses were associated with products from one specific producer
PFGE matches between patients and food products
Food histories of cases confirmed exposures
FSIS requested a recall of a lot of frozen stuffed chicken entrees
FSIS issued a letter to the industry and then labeling compliance guidance requiring label changes:
Wording changes to emphasize that the poultry is raw
Need to validate cooking instructions

19. Salmonellosis case, continued
NACMCF issued report indicating that 165o F was an effective temperature for consumers to ensure safe poultry; that microwave preparation of stuffed poultry products could be used safely
FSIS learns of new cases of salmonellosis associated with these products
FSIS issues another public health alert but no further recalls
New labels required by November 1, 2006

20. Salmonellosis case, continued
WASHINGTON, March 29, The U.S. Department of Agriculture's Food Safety and Inspection Service (FSIS) is issuing a public health alert due to illnesses from Salmonella associated with frozen, stuffed raw chicken products that may be contaminated with Salmonella.

21. Illness Investigation Objectives
Verify the association
Identify the source of production
Prevent further exposures of consumers
Initiate regulatory action (when indicated)
Identify contributing factors
Prepare an internal summary and lessons learned

22. Step 1: Human Health Surveillance
Ongoing disease surveillance conducted by
Local, State and Territorial health departments
CDC
Ongoing monitoring of foodborne disease reports by FSIS
PHELs in Omaha and Atlanta
FSIS liaison to CDC in Atlanta
Starting with Step 1, ongoing human health surveillance for disease is performed by local and state health and agriculture departments, and by the CDC and FDA.
As indicated, FSIS monitors what is going on through its PHELs and their liaison positions with the states and territories, as well as our liaison at CDC. So during this step, FSIS keeps its ear to the ground. Of course, initially we may be only identifying illness clusters and may not yet identify an actual outbreak.

23. Step 2: Illness Reported; FSIS Product Suspected
FSIS receives epidemiologic data from other public health agencies
FSIS judges the strength of the association using recognized criteria
temporal sequence; biological plausibility; specificity; dose response; relative risk/odds ratio
FSIS determines that further action is necessary
In step 2 of the FSIS procedures, the state public health officials, or CDC, are working to identify an outbreak. This work is generally supported by PFGE analysis. An epidemiologic study is launched, possibly implicating a food vehicle. During this step both FSIS and FDA are watching to see which food is implicated by the epi study.
At this point the existing data is from the States or CDC, and consists of food histories, epi studies, and sometimes lab results of food specimens. There may also be some limited traceback information such as food receipts, shoppers’ card records, etc. Often there is a report about an environmental investigation at the point of preparation or point of sale, especially when related to a restaurant.
FSIS will review the epidemiologic information provided to determine strengths of association between illness and food products.
If the analysis of all the data to this point suggests an association with an FSIS regulated product, the Agency takes further action, usually that it should investigate this outbreak and begin to generate its own data.

24. Step 3: FSIS Begins Investigation
Traceback and Traceforward
Microbiological Results
Product
Clinical
Environmental
In-plant Food Safety Assessment
During step 3, FSIS may launch an investigation focusing on tracing food back to identify a specific product and hopefully, a specific production lot and production establishment. This is a step in which information from state and local public health partners is critical to identifying the product, lot numbers, distributors, processing facilities, etc.
During a traceforward, FSIS looks from potentially contaminated product lots to distribution in other geographic areas. If there are illnesses occurring in other areas, this will factor into a request for a recall of product.
FSIS, a participant in PulseNet, also performs microbiological testing that includes PFGE analysis of isolates, such as from product and from the environment.
{NOTES:
There are a number of ways in which an in-plant Food Safety Assessment may be initiated, including:
Reports of foodborne illness, which are the most urgent; and FSIS does not need multiple cases, a single illness can trigger an assessment; also
Repetitive product sampling failures or positive microbiological testing by either the agency or the establishment, such as for E. coli O157:H7, Listeria monocytogenes, or Salmonella; and
Repetitive contamination or adulteration incidents within the plant environment.
Before we review Step 6, there are four possible outcomes of a Food Safety Assessment:
Direct evidence of adulterated product causing foodborne illness.
A preponderance of circumstantial evidence linking contaminated product with foodborne illness.
Exculpatory or no evidence.
Regulatory noncompliance; which may also occur with some of the other outcomes.} \

25. Product Traceback Investigations: What We Need to Know
When?
Production code
Sell by/use by date
Where?
Amount of product purchased
Purchase date
Point of purchase, including name and complete address
Who?
Producing establishment number
What?
Name and type
Lot number
Product code
Product weight and units per case
Percent lean
What does FSIS need to know as a public health agency to determine if regulatory action is warranted; and what would state and local health departments, and federal partners share with us, so that we can take action?
When FSIS products are implicated in an outbreak, the traceback investigations require product information. For FSIS, a traceback is literally going as far back as possible, such as from a retailer or restaurant, to the distributor, to the processor, and to the slaughterer. We need to be able to link implicated product to human illness, and then determine how it happened.
On this slide is only some of the information needed.
Who? The ID number of the plant that produced product, which is located with the USDA inspection legend.
What? The name and type of product, (such as deli meat roasted turkey breast, ground beef, ground sirloin, ready-to-eat product). It’s important to note that the packaging the product was in is just as important as the product, because of the information it contains. Also, the lot number is needed, which is a plant code for the product batch. Also, product weight units per case; and the percent lean (hamburger), such as 70%, 80%, 90%.
When? The production code, the date product was produced at the plant; and the sell by/use by date.
Where? The amount of product purchased by the consumer or case. The purchase date. The point of purchase, such as grocery store, restaurant, etc.
Important information: Is there any leftover product from the consumer or case or any other sources of the same product available for testing?
Is there any left over product held by consumer? Packaging?
Are there other sources of the same product?

26. Step 4: FSIS Analysis of Available Data
Is the epidemiologic evidence consistent with:
microbiological data?
traceback investigation?
environmental evaluation?
in-plant findings?
During Step 4, FSIS pulls together all available data from various sources to decide on further action. Is the epi data consistent with the traceback investigation? the microbiological data? the in-plant findings? and the environmental evaluations? There are ongoing consultations with public health partners, such as periodic conference calls with affected States and/or the CDC, and with Agency leadership.
Internally, epidemiologists, microbiologists, compliance investigators, and Agency leaders analyze the data. Does the information gathered or generated by FSIS fit with the epidemiologic information provided to it?
If a need for more data is identified, FSIS may go out to gather that data. FSIS may also ask for assistance from their state and local partners that are still investigating illnesses or performing product traceback within their jurisdictions, for information they may have found.

27. Here is a summary of the key findings or data that FSIS looked at with the Lm outbreak we’ve been following. This slide attempts to show which of the four processing establishments identified had problems with one or more of the four areas assessed.
First, looking at the epidemiology circle; the epidemiologic study implicated turkey deli meat. All four plants made this type of product and are therefore within the epi-circle.
Second, the traceback investigation reveals a relatively high frequency of these company’s products at the delis where the cases shopped.
Third, the micro testing revealed the outbreak strain in Plant A’s environment along with other environmental strains found in Plant A product. Plant B had the outbreak strain in 3 samples. A non-intact sample of Plant C product collected at a deli with other company’s products had the outbreak strain. Plant D had no micro positive samples and so is outside that circle.
Finally, the in-plant Food Safety Assessments revealed some problems at all four plants. Plant C is falling mostly out of the very inner area as they had good micro and sanitation programs, but some HACCP plan non-compliance that was characteristic of design-type errors that FSIS finds at a lot of plants when they go under such close scrutiny. But for Plant A we found regulatory and sanitation non-compliances. For Plant B we found adequate sanitation, but HACCP regulatory non-compliance. And for Plant D we found sanitation program problems.

28. Step 5: FSIS Regulated Product Implicated
Considerations for Agency action:
the pathogen and severity of illness
population at risk
whether new cases are being reported
strength of the epidemiologic data
strength of laboratory data
investigation at point of purchase and/or preparation
During Step 5, here are some additional factors that FSIS considers as it pulls all data and findings together. Information on what the pathogen is, the severity of illness, and the population at risk is assessed and considered. FSIS determines if new cases are being reported. Also the strength of the epi and lab data is considered. FSIS also asks about factors surrounding the preparation of the food. For example, during a restaurant outbreak FSIS would be interested in whether food handlers were ill when they were preparing food.
With the outbreak we’ve been following, FSIS considers that Lm is a pathogen that can lead to severe illness and death, and at risk populations are impacted disproportionately. The epi and lab data are strong. The possibility of deli cross-contamination is also considered. Based on all available information, FSIS identities concerns about products produced by Plants A and B.

29. Step 6: Agency Action
The Agency considers requesting a recall or detaining and seizing product in accordance with the Acts
The Agency considers a public health alert to advise about potentially adulterated product not subject to recall (i.e., not in commerce)
So during Step 6, FSIS decides on and takes further actions within their regulatory authority to protect public health. Such actions may include approaching a plant to request a recall of product. Also, FSIS may detain product in the market place or take actions to seize product. And FSIS may decide to issue alerts to consumers.

30. Step 7: FSIS Assessment of the Investigation & Lessons Learned
An internal summary documents the investigation, findings, and recommendations
Lessons learned generated by those involved in the investigation
During step 7, FSIS uses outbreak investigation findings to reassess and enhance all areas to better protect public health.

31. Recall Defined
A firm’s voluntary removal of distributed meat, poultry, or egg products from commerce when there is reason to believe they are adulterated or misbranded under the FMIA, PPIA or EPIA

32. Recall Process Problem Identification: The company finds the problem
FSIS microbiological sampling
Information from in-plant inspection program personnel
Epidemiological data gathered by Federal or State Agencies
Consumer complaints

33. Why Recall?
A Recall is a fast and effective method of removing distributed products, particularly when many lots of product have been widely distributed.

34. Who Recalls? Manufacturers and distributors of product
FSIS does not have, and is not seeking, mandatory recall authority

35. However
FSIS may initiate the recall process by informing a firm that adulterated product has been identified in commerce
FSIS retains the right to detain or seize product to protect consumers

36. www.fsis.usda.gov Food Safety and Inspection Service
United States Department of Agriculture