Transcript US Food & Drug Administration - NSTA Learning Center

LIVE INTERACTIVE LEARNING @ YOUR DESKTOP
FDA/NSTA Web Seminar:
Food Safety and Nutrition
Thursday, April 26, 2007
7:00 p.m. to 8:00 p.m. Eastern time
Foodborne Illness Outbreak Investigation
Surveillance and Epidemiology
“Be the Detective”
Patrick McCarthy, PhD, MPH
Food & Drug Administration
U.S. Dept of Health and Human Services
USDA
Food Safety and Inspection Service
– Responsible for ensuring safety of meat,
poultry, and some egg products
US Food & Drug Administration
• Responsible for ensuring safety of all other
food products
Estimates of foodborne illness and
death based on surveillance data
•
•
•
•
325,000 hospitalizations each year
5,000 deaths each year
14 million illnesses due to known pathogens
62 million illnesses due to unknown pathogens
Paul Mead-CDC-1999
Factors Pushing the Numbers Up
• Globalization of the food supply
• Perceived healthiness of raw fruits and
vegetables
• Increase in susceptible population
• Eating out more
• International travel
D Swerdlow, S Altekruse 1998
Which agency regulates these products:
Type a “U” for USDA, or an “F” for FDA
Vegetables
Chicken
Meat
Spices
Diet
Infant formula
supplements
Seafood
Pizza with a
lot of meat
Why do we do surveillance?
• Identify emerging problems and stimulate actions
to address them
• Take prompt control actions
• Identify and interpret trends in foodborne disease
• Determine the consequences of foodborne illness
• Evaluate intervention programs
• Set goals, priorities, policies, training, etc. for food
safety
How is foodborne illness recognized?
• From patients or someone close to patients
• Report from MD, RN, laboratory, etc.
• Review of national surveillance data
– Salmonella Outbreak Detection Algorithm
(SODA)
– PulseNet
– FoodNet
• Rarely: local newspaper or television news report
Percentage of Foodborne Illness
Attributable to Known Pathogens
Bacteria
30%
Protozoa
3%
Mead et al., 1999
Viruses
67%
Identify what you think is the best
reason to have a foodborne illness
surveillance program?
1.
2.
3.
Foodborne Outbreaks - FDA products only
Year
Produce
Sprouts
Dairy
Eggs
Proc.
foods
Seafood
Cosmetics
Total
1996
2
2
0
34
4
26
0
68
1997
4
3
1
32
1
8
0
49
1998
6
3
0
25
1
6
0
41
1999
9
6
0
30
3
8
0
56
2000
5
1
1
28
5
6
0
46
2001
8
3
3
16
1
13
0
44
2002
6
2
4
21
1
3
0
37
2003
6
5
4
15
3
6
0
39
2004
10
2
6
2
5
9
1
35
2005
7
0
3
4
4
6
2
26
Total
63
27
22
207
28
91
3
441
Why investigate outbreaks?
• Identify and eliminate sources of exposure
• Develop strategies to prevent future outbreaks
• Describe new diseases
• Learn more about existing diseases
• Evaluate existing prevention strategies
Why does it take so long?
Determining Burden of Disease
Reported to health department
Culture-confirmed case
Lab tests for organism
Specimen obtained
Person seeks care
Person becomes ill
Population exposures
Did you ever have a foodborne illness?
Yes ()
No (X)
If you answered yes to the previous poll
question, did you report the illness to
the health department?
Yes ()
No (X)
Multiplication Factors
For Salmonella and other pathogens that
cause non-bloody diarrhea the degree
of underreporting has been estimated at
about 38 fold.
For EC0157H7 and Shigella which cause
bloody diarrhea - underreporting has
been estimated at about 20 fold.
P Mead – CDC 1999
Burden of Illness
Measuring the Association Between Exposure and Disease
Attach Rate Table – Cohort Study
Food
Ate Food - Yes
Ate Food - No
Association
Exposure Total Ill % Ill Total Ill % Ill RR 95% CI
Oysters
19
12
63% 20
14
Salad
28
24
86% 11
2
Baked
Ham
22
16
73% 17
10
70% 0.9 0.58 -1.41
18%
4.7 1.33 -16.8
59% 1.2 0.77 -1.98
Cohort study: relative risk
total
ill
not ill
Exposed
24
4
28
Unexposed
2
9
11
•
At start of cohort study you know everyone’s
exposure status
•
total exposed = 28;
•
go forward in time and determine risk of getting ill
•
•
risk if exposed
= 24 / 28
risk if not exposed = 2 / 11
= 0.857
= 0.182
•
Relative Risk
= 4.71
total unexposed = 11
= 0.85 / 0.18
http://www-unix.oit.umass.edu/~exsc597j/Epi2000/ENGLISH/HELP/statist.htm
Take Home Message
•
In cohort studies the relative risk (RR) compares rate
of illness in the exposed group to rate of illness in the
unexposed group.
•
In case-control studies the odds ratio (OR) is the odds
in favor of exposure among cases compared to the
odds in favor of exposure among the controls.
•
if RR or OR = 1, or the 95 % CI includes 1 then the
result is not considered statistically significant
•
If RR or OR is less than 1 may mean food item is
"protective“
Review: What pathogen is associated
with the most foodborne illness ?
Bacteria
Protozoa
Viruses
Traceback
• Track food items back to their source (product
type, lot #, delivery time, etc.)
Traceback - protocol
• Traceback protocol includes
– extensive record reviews
– extensive interviews
– records / information collected includes
shipments, inventories, transportation,
etc.
– data analyses
Traceback - case study
Details:
– AZ; 19-ill; church group, restaurant, July 5th
– NV; 12-ill; friends, restaurant, July 8th
– CA; 14-ill; party, restaurant supplied food, July 9th
•
•
•
•
lab reports all cases have identical PFGE pattern
case-control study is conducted
spinach is implicated
traceback is initiated to determine source of the
spinach
Scenario - common farm
Restaurant A
Phoenix, AZ
Restaurant B
San Diego, CA
Restaurant C
Las Vegas,
NV
No common
restaurant so
distributors can be
eliminated as a
possible
contamination
source.
Spinach
Distributor A
Denver, CO
Farm B supplies
both
distributors and
is suspected
contamination
source.
Farm A
Nogales, AZ
Farm B
Salinas, CA
Spinach
Distributor B
Salinas, CA
Farm C
Tucson, AZ
Product Contamination
• 4 important sources
– soil
– water
– farm workers
– domestic and feral animals
Outbreak of Escherichia coli O157 Infections
Associated with Fresh Spinach
August-September 2006
• Early epi information
– 3 possible processors in California
– dozens of possible ranches
• Possible ongoing exposures so FDA advises
consumers not to consume bagged spinach
• Actions taken by other countries based on
early information.
Question:
Type your answers on the chat window
• Can you identify potential sources of
produce contamination in addition to
these:
–soil
–water
–farm workers
–domestic and feral animals
E. Coli O157:H7 and Spinach
• Sept. 14th; FDA notified of multi-state investigation
possibly linked to bagged spinach - possible ongoing
exposures - early epi could not identify a firm or lot
code
• Sept. 14th; California Food Emergency Response
Team dispatched to three firms
• Sept. 14th; phone calls between CDHS, FDA, and
implicated firms begin
• Sept. 15th; firm X initiated a voluntary recall
Number Of Cases by Date of Illness Onset
Number of confirmed cases
United States, August-September, 2006
30
25
20
15
August 15, 2006
Lot -227 production date
10
5
0
1
4
7
10
13
16
August
19
22
25
28
31
3
6
9
12
18
September
Date of initial symptom onset
All data is preliminary
15
21
24
Baby Spinach Harvesting
• Results:
– E.coli O157:H7 found on 4 ranches
– 9 isolates, from 1 ranch were PFGE
indistinguishable from outbreak strain
(1 stream, 1 pig feces, 7 cow feces)
• ranch is primarily a beef cattle operation
• a stream on the property - ideal habitat for wildlife
– feral pigs, etc.
• well is shallow and sits in a slight depression in
the field.
Produce Outbreaks
some lessons learned
• Leafy vegetables have elevated levels of
bacteria due to large surface area
• Bacteria tends to adhere and accumulate in
structures and at cut surfaces - once
internalized, pathogens are difficult to remove
• Pathogen survival varies greatly
• Negative lab result ≠ absence of pathogen
http://www.elluminate.com
http://learningcenter.nsta.org
National Science Teachers Association
Gerry Wheeler, Executive Director
Frank Owens, Associate Executive Director
Conferences and Programs
Al Byers, Assistant Executive Director e-Learning
NSTA Web Seminars
Flavio Mendez, Program Manager
Jeff Layman, Technical Coordinator
Susan Hurstcalderone, Volunteer Chat Moderator
LIVE INTERACTIVE LEARNING @ YOUR DESKTOP