Foodborne Outbreak Investigations What is an Outbreak An
Foodborne Outbreak Investigations
What is an Outbreak? An incident in which two or more persons have the same disease, have similar clinical features, or have the same pathogen – thus meeting the case definition – and there is a time, place or person association among these persons (CDC definition)
Foodborne Disease Outbreak (FBDO) Defined as the occurrence of two or more cases of a similar illness resulting from the ingestion of a common food (CDC definition)
Goals of FBDO Investigation Control illness and prevent future illness via: • Regulation • Promote good food handling practices among food employees and the public Provide insights into: • Agents • Food vehicles • Food handling practices
Foodborne Outbreaks, U. S. - 2007 National Totals - 1, 097 foodborne disease outbreaks (FBDOs) • Cases of foodborne illness – 21, 244 • Deaths – 18 • Bacterial – 11 • Viral – 2 • Chemical – 1 • Unknown Etiology - 4 MMWR August 13, 2010 / 59(31); 973 -979
Foodborne Outbreaks, U. S. - 2007 45% of outbreaks with a single etiologic agent 33% of outbreaks with an unknown etiologic agent MMWR August 13, 2010 / 59(31); 973 -979
Virginia FBDO Data, 2002 -2006 Year # of FBDOs Agent ID 2002 13 13 Vehicle ID Conf. /Sus. 6/2 2003 12 10 5/5 2004 18 12 5/4 2005 13 12 3/7 2006 24 21 12/4
Percent of FBDOS by Etiologic Agent, U. S. , 2007 1% 7% 40% 52% MMWR August 13, 2010 / 59(31); 973 -979
Percent of FBDOS by Etiologic Agent, Virginia, 2002 -2006 2% 44% 54%
Number of Illnesses Caused by FBDOs by Etiologic Agent, U. S. , 2007 1% 48% 1% 50% MMWR August 13, 2010 / 59(31); 973 -979
Foodborne Outbreaks, U. S. - 2007 # of Ill Persons in FBDO % of FBDOs with No Confirmed/Suspected Etiology Identified No more than 2 51 3 -7 40 8 -9 >10 30 24 MMWR August 13, 2010 / 59(31); 973 -979
Foodborne Outbreaks, U. S. - 2007 21% were attributed to a single food commodity MMWR August 13, 2010 / 59(31); 973 -979
Food Commodities CDC classifies implicated foods into 17 food commodities: • • • Finfish Crustaceans Mollusks Dairy Eggs Beef Game Pork Poultry Grains-beans Oils-sugars Fruits-nuts • Fungi • Leafy vegetables • Root vegetables • Sprouts • Vegetables from a vine or stalk MMWR August 13, 2010 / 59(31); 973 -979
Foodborne Outbreaks, U. S. 2007 Pathogen commodity pairs responsible for the most outbreak related illnesses: • Norovirus in leafy vegetables (315 illnesses) • STEC O 157: H 7 in beef (298 illnesses) • Clostridium perfringens in poultry (281 illnesses) MMWR August 13, 2010 / 59(31); 973 -979
Foodborne Outbreak Investigation Components Together Many actions are occurring simultaneously Everyone Achieves More Laboratory Epidemiology Environmental Health
Assemble Your Epi Team Communicable Disease – includes epidemiologist, public health nurses Environmental Health Administrative – includes your health director, nurse manager, support staff Lab – not usually onsite, but always available for consultation May need a core team as well as an extended team for larger, more difficult investigations
Local Health Department Preparation Define roles and responsibilities Have needed supplies on hand (DCLS supplied Foodborne Pathogen Surveillance and Outbreak Kits) Procedures for sharing information (internally and externally) Periodic meetings Training plan
Environmental Health Role in an Outbreak Investigation Capture initial complaint Environmental assessment/food preparation review of food establishment Collect food/environmental specimens Recommendations / precautionary controls
Environmental Investigation – Initial Considerations Meet with the communicable disease staff to discuss available information Laboratory Epidemiology Environmental Health • Any lab results (organism will help guide your thought process as to the contributing factor(s) involved in this outbreak) • If no lab results, determine the range of onset of illness dates among ill and see if an incubation period may be estimated • Number of folks ill / total number that attended the event/establishment
Environmental Investigation – Initial Considerations Before heading out into the field: Try to hone in on suspected exposure(s) (ex: food item) that appears to be suspicious based upon initial information from case (hypothesisgenerating) interviews Consider bringing out a member of the communicable disease staff with you on the inspection of the establishment
Environmental Investigation • Inspect the kitchen • Interview foodhandlers – similar illness? when? – procedures followed in food preparation, storage – must gain their confidence and ensure their confidentiality • Take opportunity to educate/recommend
Contributing Factors of FBDOs A confirmed organism in a FBDO investigation is helpful because it can guide you towards more probable contributing factors Improper holding temperatures Inadequate cooking Contaminated equipment Unsafe source Poor personal hygiene Cross contamination Norovirus
Communicable Disease Staff Role in an Outbreak Investigation Activate epi team Review surveillance data Literature review Questionnaire / Interview Clinical specimen collection Analysis / Report
Establish Existence of Outbreak • Outbreaks should be reported by rapid means • Information to ascertain ASAP: – How many are ill? – What are the symptoms? – When did illness begin? What is duration? – Are people still getting ill? – What is the suspected exposure? (Where? When? What? How many had same exposure? )
If Not Reported as Outbreak • • • Check Surveillance/Disease Reporting How much disease activity is normal? How much is occurring now? Are the disease reports reliable? Is this really more than expected? • Check surveillance data in VEDSS
The Surveillance Pyramid Reported to LHD State HD CDC Culture-confirmed case Lab test for organism Specimen obtained Person seeks care Person becomes ill Population exposures Source: Centers for Disease Control and Prevention
Is this really more than expected? • Look at trends • Go with your gut feeling • Get opinion of providers • May need to look at clusters in time and space • May not know much when making the determination
Start with the Sick People (Cases) • Get confirmation of the diagnosis – From physicians, labs, hospitals. • Talk to the sick people • Often start with VDH forms (ECRF or disease specific form) to record symptoms and food history • Detailed histories, including home, restaurants, events; ask about water • Assemble an investigation (Epi) team
Get Confirmation of the Diagnosis Epi-1 Form http: //www. vdh. virginia. gov/ep idemiology/documents/pdf/e pi 1. pdf
Talk to the sick people – VDH Enteric Case Report Form http: //vdhweb/epi/enteric. pdf
Find more cases Identify others who may be ill: • Get information from cases • Conduct active surveillance • Hospitals, physician practices, schools, nursing homes, restaurant etc. Need a case definition
Case Definition • Ensures consistency in who is considered ill • Use descriptive data collected in initial portion of outbreak investigation: contains elements of • • symptoms/diagnosis time place person • Changes over time. Initially loose for case finding and becomes more precise as the study progresses.
Case Definition Example: For the purposes of case identification and analysis, a case of salmonellosis was defined as a person who: Developed gastroenteritis characterized by diarrhea and at least one of the following: fever, chills, nausea, or vomiting, or laboratory confirmation of Salmonella; and Was a restaurant patron, a catered event guest, or a catered event server; and Consumed food prepared by Restaurant X on April 27 th, April 28 th or April 29 th
What do cases have in common? • Make a line list – age, race, sex, onset, duration, symptoms • Orient the data by person, place, and time. Draw an epidemic curve. • Look for similarities that came out of the case interviews. • Develop hypotheses – Based on initial data, what do you think happened?
Sample Line List
Number of Cases of Norovirus Illness by Date and Time of Onset Lunch served p. m. Feb. 19 a. m. p. m. Feb. 20 a. m. Feb. 24 p. m.
Number of Cases of Norovirus Illness by Date of Onset Point source vs person-to-person
Possible Unknowns • At this point, you might not have clear information on: – the diagnosis – the exposure • Factor in the symptoms, possible incubation period, possible exposures and try to make an educated guess. • Additional hypothesis-generating questionnaires available at www. cdc. gov/foodborneoutbreaks/standard_ques. ht m
Possible Unknowns • Recommended resource is the IAMFES book, Procedures To Investigate Foodborne Illness.
Symptoms and Incubation Period • Staphylococcus aureus – Nausea, vomiting, abdominal cramps, often diarrhea – Incubation period usually 2 -4 hours (1/2 -8) • Norovirus – Vomiting and/or diarrhea, may have fever – Incubation period usually 24 -48 hours
Collect Samples • Human • Food • Environmental • These may help you confirm the diagnosis and the exposure.
Conduct the Study • Do the interviews. • Need logistical/administrative support – people to do the interviews, phones, computers, printers, copy machines, fax machines, cars (sometimes)
Epidemiologic Investigation • Develop a standardized questionnaire based on case interviews – Pertains to specific situation under investigation • Questionnaire will be designed to test your hypotheses
Design a Study • Must have a comparison group. – Cannot determine source of the outbreak from talking to cases only. – Ill and well people will be interviewed • Can find an exposed cohort of individuals • or • Can find controls
Epidemiologic Investigation – Analytic Studies performed to test hypotheses May show an association between the exposure and illness Analytic studies include: Cohort Case-Control
Cohort Study Group of exposed and unexposed individuals Easily identifiable Compare risk of illness by what was/was not eaten Interview the entire group (ill and well)
Cohort Study Statistics Attack Rate (AR): Incidence rate in a defined population observed for a limited time, i. e. , during an outbreak Can be event-specific or food specific Used only in cohort studies Number of ill persons AR = Total population at risk, for a limited period of * 100 observation (i. e. during an outbreak)
Cohort Study Statistics Food-Specific Attack Rates (AR): AR in exposed, i. e. , people who ate each item (AR 1) • # ill who ate eggs / # who ate eggs * 100 AR in unexposed, i. e. , people who did not each item (AR 2) • # ill who did not eat eggs / # who did not eat eggs * 100
Cohort Study Statistics Relative Risk (RR): Shows the risk of developing disease given the exposure Compares the attack rate among exposed (AR 1) and unexposed (AR 2) AR 1/AR 2 = Relative Risk (RR)
Case-Control Study Used when: • you have cases but you cannot identify a group they are part of • Exposed group is very large • Compare ill with non-ill individuals to determine likelihood of having the same exposure Example: community-wide outbreak, sporadic cases of salmonellosis matching by PFGE
Case-Control Study • Comparison group is referred to as controls • Controls are free of disease and representative of population from which cases arose • Possible sources: – Identified by cases – Same neighborhood (ex: www. whitepages. com) – Same telephone exchange – Same hospital or physician
Case-Control Study Statistics Odds Ratio (OR): Measurement of the odds of having an exposure (specific food consumption) given the disease OR = Odds of exposure among cases Odds of exposure among controls
2 x 2 Tables Used to summarize counts of disease and exposure in order to do calculations of association Outcome Exposure Yes No Yes a b No c d
2 x 2 Tables a = number who are exposed and have the outcome b = number who are exposed and do not have the outcome c = number who are not exposed and have the outcome d = number who are not exposed and do not have the outcome
Measures of Association Between Exposure and Disease Cohort study RR = a / (a+b) c / (c+d) Case-control study OR = ad/bc RR/OR = 1 No Association RR/OR < 1 Negative Association RR/OR >1 Positive Association
Interpretation If the RR = 2 • People that ate Filet Mignon were 2 times as likely to develop Salmonellosis when compared with persons that did not eat it Other considerations: Validity of observed findings. . . • May be statistically significant • May be due to chance (P-value >. 05) • Could be due to bias (recall, interviewer, misclassification)
Data Analysis • Create a database • Enter the data • Analyze the data, looking for important statistical differences between ill persons and those who are not ill
Defining Roles in an Outbreak Investigation - Laboratory Laboratorians: Provide consultation regarding collection, submission and testing of specimens Perform confirmatory testing on specimens Forwards specimens to other laboratories when testing is beyond their capabilities (intestinal parasites, norovirus testing of food) Provides consultation regarding interpretation of results Assists with laboratory component of narrative report
Outbreak Investigation Concluding Actions • After analysis, interpret the results and draw conclusions • MAKE PUBLIC HEALTH RECOMMENDATIONS TO PREVENT FURTHER SPREAD OR RECURRENCE • Communicable disease staff complete the “Spoon and Fork” form and write a narrative report (may ask for your assistance for EH section) – document methods, results, recommendations made
Outbreak Investigation Concluding Actions Final meeting with epi team to tie up loose ends; consider After Action Review/Lessons Learned for larger, tougher investigations Remember to highlight the good aspects of the investigation Continue to assess the compliance of those that contributed to causing the outbreak
Additional Resources Disease Control Manual, Foodborne Outbreaks Guideline on Office of Epi Internal Web Site at http: //vdhweb/epi/food. pdf VDH Food Safety website found at www. vdh. state. va. us/Environmental. Health/Food Safety/
Acknowledgments C. Diane Woolard, Ph. D, MPH Julia M. Murphy, DVM, MS, DACVPM Chris Gordon, REHS, CFSP, MA NEHA Epi-Ready Materials
- Slides: 65