Intent of the workshop

SLE-2YEARS (1/98-1/00)Intent of the workshop
To give a functional understanding of epidemiology and it’s relevance to food safety
To discuss the current and the future uses of epidemiology by FSIS
To give you an understanding of the cascade of events that will lead to and result in, if/when you are subjected to an epi/epi-related investigation
Samadpour M, UW Environmental Health, 2000
Epi/Epi related investigation
Events leading to an investigation
Linkage of illnesses to food produced in a given plant:
Linkage by epidemiology
Linkage by epi and PFGE
Frequent isolation of a pathogen from products linked to a plant
Failing the Second/third Salmonella set
Rumors and innuendo
What is Epidemiology
Epidemiology of infectious disease
Person, place, time, infectious agent, vehicle of infection, sources of the infectious agents
Who
What
Where
When
How
Epi investigation example
In our last epi-workshop we had 100 participants. Two to five days after the workshop, 18 of the participants called and reported a sever gastrointestinal illness. The workshop faculty contacted the rest of the participants and identified 12 more cases. Based on their phone interviews of the cases and the rest of the participants the epi-investigation team determined that the cause of the outbreak was one of the items served at the lunch buffet at the workshop. Each of the participants were asked to fill out an epidemiological questionnaire.
Epi investigation example cont.
Refer to the Epi Example handout, identify the source of the outbreak and calculate the following parameters:
Incidence Rate
Mortality Rate
Relative Risk
Odds Ratio
Foodborne Outbreaks
I. Identification
Epidemiologists in at least one State Health Department notice an increase in number of cases for a reportable disease (week 1-2), or a cluster of isolates will be identified by genetic fingerprinting by one of the state health department laboratories (the PulseNet), or other subtyping methods.
Patients are interviewed and their food history is taken (week 2-3)
On the basis of the epidemiological data, exposure to a common source is determined (Week 2-4)
Foodborne Outbreaks
The CDC, FDA and USDA are notified
A case control study is conducted to determine the relative risk and odds ratio for each of the possible sources of the outbreak (Week 3-4)
All clinical microbial isolates from the recent cases are subtyped; the results will be used in conjunction with the epidemiological data to determine the outbreak cluster, and identify the source (Week 3-5)
After the source is identified, the company is notified, and public announcements are made
(Week 3-5)
Foodborne Outbreaks
II. USDA, CDC, FDA On-Site Investigation
A team of USDA and/or state health department and/or the CDC investigators will visit the plant to conduct source tracking and investigate the “How” and “Why” of the outbreak
The team will take samples (shelf-life, new products and environmental) and scrutinize production records and manufacturing practices
Foodborne Outbreaks
II. USDA, CDC, FDA On-Site Investigation Cont.
On the basis of the findings, a theory will be formulated as to the “How” and “Why” of the outbreak (Week 4-6)
The recall may be expanded based on the findings of the group
During this investigative phase, communication between the Company and the Feds will be at its maximum level
The CDC will publish their report on the outbreak (Week 20- )
Problems with epidemiological investigation of foodborne outbreaks
Most investigations are conducted properly
Too slow to be beneficial to food industry
Less shoe leather epidemiology
Chasing high profile pathogens
Scheduling the execution before the trial
Many outbreaks go undetected
Who are epidemiologists
Epidemiology of infectious diseases as applied to foodborne outbreaks is a simple and elegant science
MS or Ph.D. in epidemiology
BS, MS, Ph.D., MD, DVM, and training (formal or informal) in epidemiology
The knowledge of the food industry and food processing depends on the individual's personal or professional background, it is often lacking
Assumptions made by most epidemiologists
Until proven otherwise, for a given pathogen, the universe of possibilities for sources of the outbreak are the ones that have most frequently been associated with the previous outbreaks
For a given disease, in the absence of an increase in baseline numbers there is no outbreak
Tools used in epidemiological investigations: Data gathering and analysis
Investigations are conducted at local levels
Communications with health care providers
Reportable diseases
Active management of infectious disease reports, watch for increase in number of cases
Patient interviews, phone, epi questionnaire
Epi tools cont.: Data
Epi tools: Laboratory data
Food labs (private, county, state, federal): testing food and environmental samples
State Health Departments: reference labs to verify the isolates
State Health Department: PFGE analysis as a part of the PulseNet
Use of Molecular Epidemiology
Molecular Epidemiology
Application of the principles and methods of molecular biology immunochemistry, microbial population genetics, and epidemiology to the identification and tracking of outbreak clones.
Basic Assumptions
Clones stable within the outbreak window
Most outbreaks involve a single clone
Enough diversity at the species level that outbreak clones can be differentiated from each other and from sporadic cases
Reality
There could be more than one clone involved in an outbreak
A clone can present from multiple sources
In the absence of epidemiological linkage a molecular match does not have epidemiological significance
Clonality
Result of asexual reproduction
When a number of strains isolated at different times, from different sources, and different places have have identical phenotypic and genotypic characteristics that the only logical conclusion is that they are of clonal descent.
Clonality cont.
There is a statistical likelihood of clonality
True clonality vs. perception of clonality
Perception of clonality a function of methodology
METHODS
Genotypic methods: Plasmid profiles, restriction fragment length polymorphism based methods, ribotyping, PFGE, MRF, PCR based methods, DNA sequencing
Phenotype Based Methods
Genotype Based Methods
Fatty Acid Analysis
PCR Based Methods
The sensitivity continuum of some of the widely used subtyping methods. Phenotypic based methods (methods based on the expression of phenotypes) are at the less sensitive domain of the continuum while genotypic based methods constitute the more sensitive end of the spectrum. *The level of sensitivity depends upon the choice of gene(s) and the size of fragment(s) sequenced.
More
Sensitive
What is PulseNet? PulseNet is a national network of public health laboratories that performs DNA "fingerprinting" on bacteria that may be foodborne. The network permits rapid comparison of these "fingerprint" patterns through an electronic database at the Centers for Disease Control and Prevention (CDC). The DNA "fingerprinting" method is called pulsed-field gel electrophoresis (PFGE).
What is PulseNet? PulseNet is a national network of public health laboratories that performs DNA "fingerprinting" on bacteria that may be foodborne. The network permits rapid comparison of these "fingerprint" patterns through an electronic database at the Centers for Disease Control and Prevention (CDC).
The DNA "fingerprinting" method is called pulsed-field gel electrophoresis (PFGE).
Why was PulseNet developed? In 1993, a large outbreak of foodborne illness caused by the bacterium Escherichia coli O157:H7 occurred in the western United States. Scientists at CDC performed DNA "fingerprinting" by PFGE and determined that the strain of E. coli O157:H7 found in patients had the same PFGE pattern as the strain found in hamburger patties served at a large chain of regional fast food restaurants. Because this outbreak and its cause were recognized quickly, the ground beef patties were recalled, and an estimated 800 illnesses were prevented.
Why was PulseNet developed?
Because PFGE had such an important role in this investigation, and state health departments had increasing demands for DNA "fingerprinting," CDC developed standardized PFGE methods so that patterns from different laboratories could be generated the same way and could be compared accurately. In collaboration with the Association of State and Territorial Public Health Laboratory Directors, CDC created PulseNet so that scientists at public health laboratories throughout the country could rapidly compare the PFGE patterns of bacteria isolated from ill persons and determine whether they are similar. Similar PFGE patterns suggest that the bacteria isolated from ill persons come from a common source, for example, a widely distributed contaminated food product. Strains isolated from food products by regulatory agencies can also be compared with those isolated from ill persons. Identifying these connections can help to detect outbreaks and remove contaminated foods from the marketplace.
Limitations of genetic fingerprinting data
A two way sward
Genetic fingerprinting is not a substitute for epidemiology and/or common sense
Accurate subtyping of microbes is not an easy task, the use of multiple subtyping methods, and a high level of training for the interpretation of the data
Mistakes have been made in subtyping and application of subtyping data
Bases for epidemiological linkage of a food producer and an outbreak
Direct evidence: Very high relative risk and odds ratio, and isolation of the pathogen with exact genetic fingerprints from unopened packages of the product or from the production line
Circumstantial evidence: low to moderate relative risk and odds ratio, either no isolates from the plant or the products, or isolates which are genetically related but are not identical (human Vs. monkey)
Recall public notification
Are to be done when the source of the outbreak are positively identified, and there are reasons to believe that the contaminated batch of the product is still available in the market
Examples: a good and a bad recall
Done properly it will reduce your exposures
A not so voluntary recall: When you are approached to do so, in the absence of compelling evidence, you are left with very little choice
Resources needed when a plant is implicated in an outbreak
Same as the resources that need to be in place before it happens
A proactive approach is more cost effective
A crisis prevention/management team composed of: in-house QAQC team, production managers, physical plant management, industrial hygienist, sanitarian, public affair and legal counsel, lead by a person with authority to make decisions on the spot
The team can be augmented by outside consultants and advisors from trade associations
How to get outside help
Best resource are other companies that have been in similar situations and have survived the process; the trick is to find one!
Trade associations
Crisis management team
Ideally the team should have a clear mandate and goals
Ideally the team will work with investigative bodies (SHD, CDC, USDA, FDA) in a collaborative manner
Some of the SHDs are more open and cooperative than others, they do a good job of keeping the companies informed
Unfortunately this is not the norm
Of the federal agencies USDA is the most accessible as far as sharing the available information
Crisis management goals
Fact finding: Is this our outbreak, if yes: how to help the effected individuals, how to contain the outbreak, how to resolve the underlying problem(s), how to minimize the impact on our customers, how to manage our employee’s moral, how to respond to media inquiries, how to respond to consumer inquiries, how to respond to document request, how to manage the recall, how to deal with our vendors, how to deal with our insurance providers, legal issues, potential criminal charges, implications for our brand names implicated, implications for our other brand names, etc., etc.
Crisis management goals cont.
Is this our outbreak: If the linkage is based on circumstantial evidence then:
Aggressive auditing of the epidemiological investigation, and the laboratory data
Maintain open channels of communications w/FSIS
Let the science be your defense
You are entitled to the results of the investigations as they become available
Analyze the data and continue an active discussion with those involved in the investigation
This may come as a shock to them, the reaction ranges from indignation to respect
Foodborne infectious disease outbreaks are easy to detect and document
Lack of clear evidence is the best indication that a company may not be involved in an outbreak
Some epidemiologists jump to conclusions in the absence of clear evidence and contrary to the scientific tradition
Till recently the FSIS had not taken action (issuance of NR, IDV, epi investigation, etc.) in the absence of a positive culture isolated from an unopened package of product by a sanctioned lab
Preparing for the swat/epi/idv team
Anywhere between the first week of the outbreak/recall to months after, the plant may be subjected to IDV/epi review
Any food production plant any day, anytime should be prepared for an unexpected visit
It is prudent for any production plant to undergo IDV audits
Aside from preparing in advance for such visits, have the crisis management team with appropriate experts and consultants be on site at the time of the IDV/epi review
The IDV/epi review team
The team will pour over all the available pertinent data regarding your production, HACCP, HACCP implementation, CCPs, NRs, intervention steps, laboratory data, shipping records, vendor and customer information, etc, etc,
They will conduct several inspections of the plant at different times
They will interview your personnel, and will consult with the USDA personnel at the plant and at the District Office
Some of the team members are USDA professionals with years of experience, and some have little experience with food production plants
IDV/epi review process
Initial interview: establish a partnership, negotiate your expectations, set milestones, ask for daily briefing
Have your expert/legal team onsite
Be cordial, be firm, be on your guards, it is a cross between a physical exam and tax audit for a middle aged person of some means (including high sugar and cholesterol)
They are there to find problems, and problems they will find, in a well run plant there should not be major problems, your experts should have identified and rectified all the problems beforehand
IDV/epi review process cont.
Sampling: The team should not have to sample. Sampling by the team could automatically put you on test and hold
Do not volunteer the resources
Daily progress meeting: Discuss their findings, involve your experts, correct the problems as they are found by either side (your consultants or the IDV team)
Ideally the team will present a balanced view of the production practices and the plant, they will identify areas for improvement, and make recommendations accordingly
If you disagree with their findings be prepared for immediate challenge at the agency or legal level
Possible outcomes of IDV/epi review
Recommendation for plant closure: Fight
Expanded recalls: In the absence of public health problems: Fight
Identification of minor issues and concerns: Within a month you will receive a report issued by the team, you will have a month to respond, correct all the problems that they have identified, or you have become aware of during your own audits, this is the main benefit of such reviews
Prepare and submit your response
Transition from crisis management to defense
After the first month or so the public interest in the outbreak will begin to decline
The epidemiological investigation will often continue for a prolonged period
The IDV/epi review may have ended by now
The official end of the investigation is issuance of the Epi Report by the SHD in charge or the CDC
While the public relations and the customer management issues will continue, the legal team will have to start with a large number of FOIA requests to appropriate agencies, continue with auditing the outbreak
Foodborne Outbreaks
III. The Press
In states which have cases, the outbreak will be the top story in the news media
The amount of media play at the national level will depend on the extent of the outbreak, and the organism involved
Patients, their families, relatives of the deceased, and current and former employees of the company will be interviewed
All inspection records, previous recalls, and health and safety violations will be publicized
On the basis of the Company’s testing data, a “Smoking Gun” will be created
Foodborne Outbreaks
IV. Lawsuits: A) USDJ
There has been an increased tendency on the part of the USDJ, to bring criminal charges against the companies involved in large/high profile outbreaks
Criminal charges have been brought against Odwalla, Hudson and Sara Lee
Foodborne Outbreaks
IV. Lawsuits: B) Class Action
One recent foodborne outbreak has resulted in a class action lawsuit representing the patients
Another major foodborne outbreak resulted in a CALS by franchise owners
Foodborne Outbreaks
Large numbers of cases are filed
Some are unrelated to the outbreak
While a few law firms are experienced in the area of foodborne diseases, most law firms on both sides have no prior experience in this area
It is best if cases are settled as fast and as early as possible
Foodborne Outbreaks
All microbiological testing data
All memos, e-mails, correspondences between the employees, the company and customers, regulators, consultants, contractors, auditors, etc.
All records of previous problems at the plant
Foodborne Outbreaks
IV. Lawsuits: D) the Press
Successful attorneys will hold public interest and use the media to their favor by keeping the story on the front page as long as possible
Everyone will talk to the press except the Company officials and their attorneys
Foodborne Outbreaks
Full Cooperation with the Feds
Agreement to recall
Some level of support is provided for patients
There is a precipitous drop in the market share
There is a precipitous drop in market capitalization
Foodborne Outbreaks
Negotiations with the USDA
Bankruptcy
Foodborne Outbreaks
Acting guilty
becoming non-communicative
Sudden urge to fund research in food safety
A desire to settle cases and get them over with ASAP
Foodborne Outbreaks
Better surveillance and epidemiology
The nude mice syndrome
Foodborne Outbreaks
Foodborne Outbreaks
“The plant was as good as it gets”
“We tested our products”
“The contaminated raw material was from another producer”
Foodborne Outbreaks
Take your check book
Foodborne Outbreaks
Foodborne Outbreaks
Prevention? Cont.
Plant Design
Testing, testing, testing!
Foodborne Outbreaks
Release of contaminated products, population exposure, clinical cases, epidemiological linkage, identification of the source
Competition between the public health system and the food industry
The PHS is getting better
Foodborne Outbreaks
Why Test?
To search for pathogens
To determine the quality of raw and finished products
Shelf-life studies
Foodborne Outbreaks
Testing: How?
Five Models:
Decentralized
Centralized
Plant labs for routine monitoring and a central lab for specialized testing
Spin off the QA/Lab Services Division into an independent company
Foodborne Outbreaks
Should be pathogen FREE
Require a minimum of B.S. Degree for the plant lab techs
Should test for general indicators of microbiological quality of the products and the environment
Foodborne Outbreaks
Testing: The Central Lab
At least the same level of competency as the CDC/USDA/FDA lab
Should be able to perform the same tests and maintain the same database as the Federal labs
These include:
testing for specific pathogens using the classical micro methods, plus the use of molecular biology based and immunochemical based methods
Methods in molecular epidemiology
Foodborne Outbreaks
Local Vs. Central Testing
Impossible to keep all the expertise needed in every plant lab
QAQC issues become draconian
Foodborne Outbreaks
Food Safety Chain of Command
Learn from our government: USDA, FDA, CDC, Department of Commerce, Human and Health Services, State Health Department, State Department of Agriculture, County Health Departments, City Health Departments, etc., etc.
There should be a clear, concise and independent chain of command
Food safety issues should supercede production issues
What is PulseNet?
performs DNA "fingerprinting"
foodborne. The network permits
rapid comparison of these
"fingerprint" patterns through an
for Disease Control and
Prevention (CDC). The DNA
"fingerprinting" method is called
public health laboratories that performs
DNA "fingerprinting" on bacteria that
may be foodborne. The network permits
rapid comparison of these "fingerprint"
patterns through an electronic database
at the Centers for Disease Control and
Prevention (CDC).
called pulsed-field gel electrophoresis
foodborne illness caused by the
bacterium
Scientists at CDC performed DNA
"fingerprinting" by PFGE and
E. coli
hamburger patties served at a large
chain of regional fast food restaurants.
Because this outbreak and its cause were
recognized quickly, the ground beef
patties were recalled, and an estimated
800 illnesses were prevented.
investigation, and state
health departments had
increasing demands for

Intent of the workshop

Documents