CDC Epidemic Intelligence Service Officer Tennessee Department of Health Council of State and Territorial Epidemiologists Annual Meeting June 13 2011 Characteristics of Foodborne Disease Outbreak Investigations Conducted by FoodNet Sites 20032008 ID: 313543
Download Presentation The PPT/PDF document "Rendi Murphree, PhD" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Rendi Murphree, PhD
CDC Epidemic Intelligence Service OfficerTennessee Department of HealthCouncil of State and Territorial Epidemiologists Annual MeetingJune 13, 2011
Characteristics of Foodborne Disease Outbreak Investigations Conducted by FoodNet Sites, 2003–2008
Office of Surveillance, Epidemiology, and Laboratory Services
Scientific Education and Professional Development Program Office Slide2
BACKGROUND
FoodNet Outbreak InvestigationsSlide3
Annual Burden of Foodborne Illness and Outbreaks in the United States
Foodborne illnesses (estimated)9.4 million illnesses
1,351 deaths31 major pathogens
Foodborne outbreaks (reported)20,000 illnesses
20 deathsInvestigations identify factors contributing to foodborne illness
Etiology and food vehicle often unknown
Scallan et al.,
Foodborne illness acquired in the United States
, Emerg Infect Dis, 2011, Jan 7(1)7-15
CDC, Foodborne Outbreak Online Database,
http://wwwn.cdc.gov/foodborneoutbreaks
Accessed 28 June 2010. Slide4
Foodborne Disease OutbreakSurveillance Systems
CDC Foodborne Disease Outbreak Surveillance SystemNORS (National Outbreak Reporting System)All statesOutbreak characteristics
Foodborne Disease Active Surveillance Network FoodNet10 states, CDC, FDA and USDA~45 million people; 15% of the U.S. population
Supplemental data on investigation characteristicsSlide5
Foodborne Disease OutbreakSurveillance Systems
CDC Foodborne Disease Outbreak Surveillance SystemNORSAll states
Foodborne Disease Active Surveillance Network FoodNet10 statesSlide6
OBJECTIVES
FoodNet Outbreak InvestigationsSlide7
Study Objectives
Summarize data on FoodNet outbreak investigations
Determine factors associated with identifying an etiology and implicating a food vehicle
Describe barriers to successful investigations
Recommend investigation improvementsSlide8
METHODS
FoodNet Outbreak InvestigationsSlide9
Data Source
FoodNet sitesCT, GA, MD, MN, OR, TN
Counties in CA, CO, NYNM since 2004
2003–2008
Multi-state outbreaks excluded
Combined NORS and FoodNet dataSlide10
Variables
NORS FormInformation on the outbreak
FoodNet Outbreak Supplement Form
Information on the outbreak investigationRecognition and reporting
Investigation designSpecimen testingBarriers impeding the investigationSlide11
Definitions
Foodborne disease outbreak (FBDO) ≥2 cases of a similar illness resulting from the ingestion of a common food
Food vehicle identified
Statistical evidence or laboratory evidence
Etiology confirmedCDC criteria
Odds ratios adjusted for outbreak size
*Olsen, et al., MMWR Surveillance Summary, 2000;49(1):1-62Slide12
RESULTS
FoodNet Outbreak InvestigationsSlide13
FBDOs — FoodNet Sites, 2003–2008
42 multi-state outbreaks excluded
1,200 outbreaks analyzedAnnual mean number = 200 (range 117–253)
18,593 illness caused by 1,111 outbreaks
Mean = 17 per outbreak (range 2–273)Marked variability in outbreak frequency and investigation characteristicsSlide14
FBDO by Year — FoodNet Sites, 2003–2008
N = 1,200Slide15
FBDO Characteristics — FoodNet Sites, 2003–2008
All 10 Sites
Number reported
1,200
Average annual rate per million
4
Confirmed etiology (%)
50
Food vehicle identified (%)
32
Etiology & food vehicle identified (%)
20Slide16
FBDO Characteristics — FoodNet Sites, 2003–2008
All 10 Sites
Range
Number reported
1,200
11–279
Average annual rate per million
4
1–9
Confirmed etiology (%)
50
37–82
Food vehicle identified (%)
32
16–50
Etiology & food vehicle identified (%)
20
7–39Slide17
FBDO Seasonality by Confirmed Etiology — FoodNet Sites, 2003–2008 (n = 715)Slide18
FBDO Seasonality by Confirmed Etiology — FoodNet Sites, 2003–2008 (n = 715)Slide19
FBDO Seasonality by Confirmed Etiology — FoodNet Sites, 2003–2008 (n = 715)Slide20
FBDO Seasonality by Confirmed Etiology — FoodNet Sites, 2003–2008 (n = 715)Slide21
Stool Key to Confirming an Etiology
Stool specimens obtained in 89% of 715 outbreaks with a confirmed etiology vs. 48% of 485 withoutOdds ratio = 9.2 (6.8–12.3)Slide22
Stool obtained in 89% of 715 outbreaks with a confirmed etiology vs. 48% of 485 without
Odds ratio = 9.2 (6.8–12.3)Odds of confirming an etiology did not improve with the collection of
>4 stool specimensOdds ratio = 7.2 (4.9–10.6)
Stool Key to Confirming an EtiologySlide23
Stool obtained in 89% of 715 outbreaks with a confirmed etiology vs. 48% of 485 without
Odds ratio = 9.2 (6.8–12.3)Odds of confirming an etiology did not improve with the collection of
>4 stool specimensOdds ratio = 7.6 (5.3–10.9)
Proportion of outbreaks with a confirmed etiology high, regardless of interval between symptom onset and stool collection
Median 0–3 days (70%), 4–7 days (78%), 8–14 days (69%)
Stool Key to Confirming an EtiologySlide24
Stool obtained in 89% of 715 outbreaks with a confirmed etiology vs. 48% of 485 without
Odds ratio = 9.2 (6.8–12.3)Odds of confirming an etiology did not improve with the collection of
>4 stool specimensOdds ratio = 7.6 (5.3–10.9)
Proportion of outbreaks with a confirmed etiology high, regardless of interval between symptom onset and stool collection
Median 0–3 days (70%), 4–7 days (78%), 8–14 days (69%)
Stool Key to Confirming an EtiologySlide25
FBDO Characteristics by Outbreak Size — FoodNet Sites, 2003–2008Slide26
Keys to Implicating a Food Vehicle
Food vehicle implicated in 47% of 671 outbreaks investigated using a case-control or cohort study vs. 14% of 529 using other methodsOdds ratio = 4.9 (3.6–6.7)Slide27
Keys to Implicating a Food Vehicle
Food vehicle implicated in 47% of 671 outbreaks investigated using a case-control or cohort study vs. 14% of 529 using other methods
Odds ratio = 4.9 (3.6–6.7)
Analytic study conducted in only 20% of outbreaks causing 2–4 illnessesSlide28
Keys to Implicating a Food Vehicle
Food vehicle implicated in 47% of 671 outbreaks investigated using a case-control or cohort study vs. 14% of 529 using other methods
Odds ratio = 4.9 (3.6–6.7)
Analytic study conducted in only 20% of outbreaks causing 2–4 illnesses
Food specimens collected in 91 (36%) of 253 investigations successful in identifying a food vehicleOdds ratio
= 4.9 (3.6–6.7)Slide29
Source of Initial FBDO Recognition —
FoodNet Sites, 2003–2008
Viral
n = 471
Bacterial
n = 246
No.
(%)
No.
(%)
Private citizen
363
(87)
80
(33)
Reportable disease surveillance
11
( 3)
75
(30)
Medical professional report
34
( 8)
43
(17)
DNA match
0
( 0)
33
(13)
Syndromic
surveillance
0
( 0)
3
( 1)
Other
10
( 2)
12
( 5)Slide30
Source of Initial FBDO Recognition —
FoodNet Sites, 2003–2008
Viral
n = 471
Bacterial
n = 246
No.
(%)
No.
(%)
Private citizen
363
(87)
80
(33)
Reportable disease surveillance
11
( 3)
75
(30)
Medical professional report
34
( 8)
43
(17)
DNA match
0
( 0)
33
(13)
Syndromic
surveillance
0
( 0)
3
( 1)
Other
10
( 2)
12
( 5)Slide31
Source of Initial FBDO Recognition —
FoodNet Sites, 2003–2008
Viral
n = 471
Bacterial
n = 246
No.
(%)
No.
(%)
Private citizen
363
(87)
80
(33)
Reportable disease surveillance
11
( 3)
75
(30)
Medical professional report
34
( 8)
43
(17)
DNA match
0
( 0)
33
(13)
Syndromic
surveillance
0
( 0)
3
( 1)
Other
10
( 2)
12
( 5)Slide32
n = 370
Receipt of Initial FBDO Recognition—
FoodNet Sites, 2006–2007Slide33
Settings of FBDO —
FoodNet Sites, 2003–2008Site of food preparation (n = 1,088)
64% restaurant or deli 7% caterer
11% homeSlide34
Preparation Leading to FBDO —
FoodNet Sites, 2008Timing of contamination (n = 129)
74% at the time of preparation or serving26% prior to final preparation
Contaminated food served (n = 184)
86% single establishment or eventSlide35
Agencies Involved in FBDO Investigations — FoodNet Sites, 2003–2008
37% local health departments only
47% local and state agencies
5% local, state and federal agencies
Local health departments not involved in 12%
n = 1,185Slide36
Barriers Affecting FBDO Investigations — FoodNet Sites, 2003–2008
Notification18% delayed notification of local health departments
Cases26% too few cases
24% lack of cooperation from casesControls23% too few controls
10% inability to identify good controls for an analytic studyStool
25% too few stool specimens
n = 655Slide37
DISCUSSION
FoodNet Outbreak InvestigationsSlide38
Overview of Major Findings
FBDOs:Mainly recognized and handled locally
Attributed to exposures outside the home Caused by contamination at time of preparation
Stool specimens key to confirming an etiology
At least 4Stool collection weeks after symptom onset useful
Analytic study best for implicating a food vehicle
Cohort or case-control study
Food specimen collection rare but helpful in bacterial outbreaksSlide39
Study Limitations
Data qualityIncomplete data submitted to CDC
Questions changed over timeSubjective interpretation
No ongoing training for data collectorsSome data not reported by staff directly involved in the outbreak investigation
ResultsMay not be generalizable to non-FoodNet sitesSlide40
Conclusions & Recommendations
Public health agencies must improve investigation of foodborne disease outbreaks
RecommendationsSystematically review complaints from the public
Increase stool specimen collectionHealth departments
Clinicians Consider new methods for identifying and contacting patients and control subjects
Provide training
Conducting epidemiologic studies
Collecting NORS and FoodNet dataSlide41
Acknowledgements
FoodNet Partners10 states
CDC, FDA, USDA
EIS Supervisory TeamTimothy F. Jones
John R. DunnWilliam Schaffner
W. Randolph Daley
Coauthors
Katie Garman
Quyen
Phan
Karen
Everstine
L. Hannah Gould
Timothy F. Jones
Office of Surveillance, Epidemiology, and Laboratory Services
Scientific Education and Professional Development Program Office