making decisions despite uncertainty: the irish dioxin crisis 2008
DESCRIPTION
12 th Annual Joint FERA/JIFSAN Symposium Maryland, June 2011. Making Decisions Despite Uncertainty: The Irish dioxin crisis 2008. Dr Wayne Anderson Food Safety Authority of Ireland. Ireland: December 2008. Dioxins and e arly c risis events Areas of uncertainty and exposure assessment - PowerPoint PPT PresentationTRANSCRIPT
© 2005
Making Decisions Despite Uncertainty:The Irish dioxin crisis 2008
Dr Wayne AndersonFood Safety Authority of Ireland
12th Annual Joint FERA/JIFSAN SymposiumMaryland, June 2011
© 2005
Ireland: December 2008
• Dioxins and early crisis events• Areas of uncertainty and exposure assessment• Risk communication
© 2005
PCDD/PCDFs
The term “dioxin” covers a group of chemically similar substances:
75 polychlorinated dibenzo-p-dioxins (PCDDs) and
135 polychlorinated dibenzofurans (PCDFs) 17 of toxicological concern
© 2005
National Residues Monitoring Programme
How Did We Find It?
Pesticide Control Service, DAFF
© 2005
Time Line 2008Sunday Monday Tuesday Wednesday Thursday Friday Saturday
1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30
Sunday Monday Tuesday Wednesday Thursday Friday Saturday
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31
Pork fat sample taken
Preliminary Marker PCB result
Confirm marker PCB resultRestriction of pig movementCrumb feed sample mPCB positive
Visit to index farm
Dioxins confirmedHigh level meetingsFull recall of Irish pork
Further pig restrictions / Press statement Dutch information provided
November
December
© 2005
Decision: Sat 6th December 2008
© 2005
0
10
20
30
40
50
60
Dioxin congener
% to
tal
Series1Series2Series3Series4
Cooperation: Pattern of dioxins (absolute)
0
10
2030
40
50
60
2,3,7,
8-TCDF
1,2,3,
7,8-Pe
CDF
2,3,4,
7,8-Pe
CDF
1,2,3,
4,7,8-
HxCDF
1,2,3,
6,7,8-
HxCDF
2,3,4,
6,7,8-
HxCDF
1,2,3,
7,8,9-
HxCDF
1,2,3,
4,6,7,
8-HpC
DF
1,2,3,
4,7,8,
9-HpC
DF
OCDF
2,3,7,
8-TCDD
1,2,3,
7,8-Pe
CDD
1,2,3,
4,7,8-
HxCDD
1,2,3,
6,7,8-
HxCDD
1,2,3,
7,8,9-
HxCDD
1,2,3,
4,6,7,
8-HpC
DD
OCDD
frac
tion
of t
otal
(%
)
Irish 1Irish 2Irish 3Irish 4Irish 5Irish 6Irish 7
Irish Samples
Dutch Samples
© 2005
Science can help focus investigations
PCB profile for crumb and pig fat samples from Ireland compared to Aroclor 1260
0
5
10
15
20
25
30
35P
CB
18
PC
B 2
8
PC
B 3
1
PC
B 4
7
PC
B 4
9
PC
B 5
1
PC
B 5
2
PC
B 7
7
PC
B 8
1
PC
B 9
9
PC
B 1
01
PC
B 1
05
PC
B 1
14
PC
B 1
18
PC
B 1
23
PC
B 1
26
PC
B 1
28
PC
B 1
38
PC
B 1
53
PC
B 1
56
PC
B 1
57
PC
B 1
67
PC
B 1
69
PC
B 1
80
PC
B 1
89
% w
t
%wt Pig fat %wt Crumb % wt (Aroclor 1260) Frame et al, 1996
© 2005
Data summary
• Same dioxin and PCB profiles in pork meat samples in IRL / NL / FR
• NDL-PCBs in pork 500-3000ppb
• Ratio NDL-PCBs / dioxin-TEQ was low compared to previous incidents like Belgium
• Dioxins almost exclusively PCDFs
• Data suggests – Aroclor 1260 contamination (transformer oil)
© 2005
Main Areas of Uncertainty for Exposure
• U1: Percentage of pig herd exposed to feed
• U2: Time exposure to contaminated feed and pork
• U3: Subsequent level of contamination in pork fat
• U4: Consumption of pork and pork products in
Ireland
© 2005
U1:Percentage of the Pig Herd
10 Pig ProductionFarms
= 8% NationalPork Output
One Recycling Plant
© 2005
Why Recall Everything?
10 Major Processing Plants
98% PorkOutput
150,000 t/year
© 2005
U2: Feed Exposure PeriodCrumb Screening Results 13 Aug – 3 Dec, 2008
13-A
ug
20-A
ug
27-A
ug
03-S
ep
10-S
ep
17-S
ep
24-S
ep
01-O
ct
08-O
ct
15-O
ct
22-O
ct
29-O
ct
05-N
ov
12-N
ov
19-N
ov
26-N
ov
03-D
ec
Negative
Positive ++
Trace
Positive--
Positive
© 2005
U2: exposure period for PorkDioxin levels at rendering plant in Belgium
Exposure period 1st September to 6th December
0
0,5
1
1,5
2
2,5
3
3,5
4
31-jul 20-aug 9-sep 29-sep 19-oct 8-nov 28-nov 18-dec
pg/g
fat
© 2005
U3: Dioxins Levels in Feed and Pork Fat
• Crumb Product: E.U. Limit Feed 0.75pg/g
• Levels detected in Feed: 5200pg/g
• Pig Fat: E.U. Legal Limit 1pg/g
• Levels detected: 80 – 200 pg/g
© 2005
U4: Pork Fat Consumption Data
4% fat
Pork Casserole with Potatoes
Pork Fat 0.98%
Old potatoes, average, raw
Onions, raw
24.4%
4.8%
3.5%
2.2%
0.2%
15.3%
43.1%
6.5%
Carrots, old, raw
Pork, diced, raw, lean
Parsnip, raw
Turnip, raw
Stock cubes, chicken
Water, distilled
• Food Consumption Survey (http://www.iuna.net/) - 7 day dietary records of 958 adults, aged 18-65, from Republic of Ireland during 1997-1999
• Database containing information for each individual and each eating occasion – split into ingredients for purposes of Total Diet Study – further manipulated for the purposes of estimating exposure to lipophilic substances (i.e. POPs)
© 2005
Exposure Assessment
Food intake x Presence x Chemical = Exposure probability concentration
Values in 10^1
0.00
0.03
0.05
4.56 5.65 6.74 7.83 8.91 10.00 Values in 10^1
0.00
0.02
0.05
1.5 2.9 4.3 5.8 7.2 8.6
Values in 10^1
0.00
0.03
0.05
4.56 5.65 6.74 7.83 8.91 10.00
yesno
• Databases uploaded into commercial probabilistic software program Crème Food
www.cremesoftware.com
© 2005
FSAI Probabilistic modelling of exposure
• Total Population Intakes of Total WHO TEQ from all sources (including ingredients) assuming that 10% of the pork consumed contains a level of between 80-200 pg WHO TEQ/g fat and 90% of pork consumed contain usual background levels (as determined in previous surveys)
• All other intakes are calculated based on background levels determined in previous surveys. All intake calculations are based on ranges of data and the results presented are based on a run of 300 iterations
FSAI Probabilistic Exposure AssessmentTotal TEQ pg/kg bw/day (upper-bound)
Background exposure Background & Incident ExposureFood Group Mean P97.5 Mean P97.5Total Exposure 0.40 1.62 2.13 11.14Pork Only 0.03 0.12 1.58 10.56
© 2005
EFSA POINT ESTIMATE INTAKE FOR PORK
CountryAverage fat intake (g/kg
bw/d)
TEQ exposure (pg/day per kg.b.w.)
50 pg TEQ/g fat 100 pg TEQ/g fat 200 pg/g fat
% fat from contaminated pork % fat from contaminated pork % fat from contaminated pork
100% 10% 1% 100% 10% 1% 100% 10% 1%
Ireland 0.25 12.3 1.2 0.1 24.6 2.5 0.2 49.3 4.9 0.5
Country97.5% fat
intake (g/kg bw/d)
TEQ exposure (pg/day per kg.b.w.)
50 pg TEQ/g fat 100 pg TEQ/g fat 200 pg/g fat
% fat from contaminated pork % fat from contaminated pork % fat from contaminated pork
100% 10% 1% 100% 10% 1% 100% 10% 1%
Ireland 0.44 22 2.2 0.2 44 4.4 0.4 88 8.8 0.9
EFSA deterministic modelling of exposure
© 2005
EFSA Pork risk assessment
• Uncertainty in exposure estimate- 10% of pork contaminated- 90 day exposure- 200pg/g dioxin
• Conclusion 10% increase in body burden No concern to human health from this single exposure
event
© 2005
It’s About Food Safety…
Risk Assessment Risk Assessment
Agreement
© 2005
• 2002 breast milk study, which has a mean of 11.9 pg/g fat Assuming 60 kg body weight and 20% fat content, this gives
an estimated body burden of 2.4 ng/kg over the 4 Irish populations studied
Lower than the 4ng/kg European average
• 2010 breast milk study (pooled samples 109 first time mothers) Dioxin levels down ~20% No appreciable exposure impact of the 2008 dioxin crisis Publication submitted to Chemisphere
Body Burden of Dioxins in Ireland
© 2005
Communicating the message
We Have Identified Contamination
We Have Recalled Product
We Are Isolating the Cause
We Will Keep You Informed
Simple
© 2005
It Can Sink in ……
No Matter How Strong the Message…
© 2005
Maybe Not Simple…
385 Articles in National Press
200 Articles in Regional Press
70 Radio Programmes + phone-in’s
17 Television Programmes
200 Internet News Items
800+ Journalists Delivering the
Message?
© 2005
Did They Deliver The Simple Message?
Competition in the “news” media
© 2005
Advice Line Calls(3,725)
Calls to advice-line on pork recall
40
602
303120
2,660
0
500
1000
1500
2000
2500
3000
Saturday 6th Sunday 7th Monday 8th Tuesday 9th Wednesday 10th
Website traffic: up 4,310%
© 2005
© 2005
© 2005
But in any crisis co-operation is vital…
FERA, York RIKILT, NL VWA, NL Food Standards Agency (NI and London) European Commission European Food Safety Authority
© FSAI
Particular thanks to Christina Tlustos and Rhodri Evans for helping in the preparation of this talk