Risk assessment: Purpose and Application

1
Risk assessmentPurpose and Application

• David Vose Consultancy
• 24400 Les Lèches
• Dordogne
• France
• www.risk-modelling.com

2
What is Risk Assessment?

• R.A. should be a decision-focused exercise.
  Determine
• How big the risk issue is
• What it makes sense to do to manage the risk
• What knowledge we have to produce a reasoned
  argument between plausible risk management
  options
• What data would allow us to consider and compare
  more options
• Or compare options more robustly
• It should not be a model focused exercise
• Not building a risk assessment model of the
  process, then seeing if it answers any decision
  questions

3
Questions for Food Safety Commission to ask

• What is the amount of disease caused by food?
• What actions could be taken to optimise food
  safety objectives?
• What connections are there to risks managed by
  other agencies?
• How, and with whom, do we communicate?

• Scope of the problem (just food, or risks from
  food production?)
• Number of cases impact for a case -gt total
  disease burden
• Risk attribution Domestic, imported food,
  environment, human-to-human, etc
• Risk factors -gt clue to possible management
  options

• Change in operations that maximise human health
  benefit
• So look at effect over portfolio of risks
• Might involve removing expensive, ineffective
  standards
• Need to consider reactions to change in the world
  (economic, social)
• Look at secondary risks
• RANK risk management options, use risk registers

• Eg chemicals used in crops food risk
  environmental risk
• Eg bacteria from food-producing animal meat,
  vegetable, environment risk

• Partnership v regulation, eg Danish elimination
  of use of growth promotants
• Who has data?
• Risk managers and risk assessors making a team

4
The goal of risk management
Minimise S food related health impact

illnesses
5
Codex Alimentarius CommissionFAO/WHO (1995)

• Microbial risk assessment is a scientifically-base
  d process consisting of four steps
• Hazard Identification The identification of known
  or potential health affects associated with a
  particular agent
• Exposure Assessment The qualitative and/or
  quantitative evaluation of the degree of intake
  likely to occur
• Hazard Characterization The quantitative and/or
  qualitative evaluation of the nature of the
  adverse effects associated with biological,
  chemical and physical agents that may be present
  in food For biological agents a dose-response
  assessment should be performed if the data are
  available
• Risk Characterization Integration of Hazard
  Identification, Hazard Characterization and
  Exposure Assessment into an estimation of the
  adverse effects likely to occur in a given
  population, including attendant uncertainties.

6
Codex based on NAS chemical risk guidelines

• Microbial risk assessment has many similar
  components
• Hazard can be identified (bacterium cf compound)
• Complex exposure pathways
• Exposure from multiple sources (need for risk
  attribution)
• Dose-response relationship
• But also has differences
• Cannot usually identify causative agent for
  chemical r.a. (could be several)
• Animal testing for chemicals, species-to-species
  dose-response issue (10 fingers)
• Single-hit (microbial) v chronic (chemical)
  exposure
• Pathogens are discrete (eg detection
  difficulties)
• Pathogens grow and die, need to be cultured for
  detection
• Usually surface contamination, so concentration
  not ideal measure
• Pathogens naturally occur in environment and
  animal gut, not a contaminant
• Pathogens can mutate (resistance issues)
• Single pathogen can cause infection, no threshold

7
What have we done with these CODEX guidelines?
8
Current microbial risk assessment

• Microbial QRA is a developing discipline
• Mostly producing farm-to-fork (F2F)
• Model the whole system but very poorly
• Not designed to model any decision question well
• One pathogen in one food commodity directly
  consumed
• Often rely on poor data, surrogates, and guesses
• Almost never is a decision question posed
  beforehand
• Assessors have over-sold F2F QRAs usefulness
• Managers have expected too much
• Focused on there being exposure and D-R models
  (eg see WHO guidelines and developed models)

9
Dutch observations on past QRAHavelaar, Jansen
(2002)

• The lessons learnt from risk analysis
  experiences
• Risk management has not always been an integral
  part of risk analysis so far
• Risk managers should be trained to understand
  risk assessment, and risk assessors should be
  trained to explain their work
• Available data are often of limited use for risk
  assessment and communication of data needs
  between risk assessors, food scientists and risk
  managers is a critical issue
• The risk manager questions usually require rapid
  results, whereas (farm-to-fork) risk assessment
  projects require several years to complete.
  Solving this conflict requires open
  communication
• Uncertainty is often large.

10
Completion times of some farm-to-fork QRAs
Final report
Final report
Draft report
Being revised
Draft report
Final report
11
USDA-FSIS-FDA Salmonella Enteritidis

• Although the goal was to make the model
  comprehensive, it has some important limitations.
  . For many variables, data were limited or
  nonexistent. Some obvious sources of
  contamination, were not included. And, as
  complex as the model is, it still represents a
  simplistic view of the entire farm-to-table
  continuum. Much more work is needed to address
  all limitations.

12
Conflict between microbiology and risk assessment
focus(Adapted from presentation by Maarten
Nauta, 2002)

• Microbiology
• About the detectable
• Micro-world
• Selected strains
• Against variability
• Qualitative
• Science

• Risk Assessment
• What is there (prevalence, load)
• Macro-world
• All strains
• Pro variability
• Quantitative
• Decision tool

• Risk assessors and microbiologists need to work
  more closely together. Microbiologists need to
  learn new methods of procedure and reporting.

13
Where do we go wrong?Separation of managers and
assessors

• CODEX recommended the separation of the roles of
  managers and assessors
• To allow science to evaluate risks unhindered by
  politics
• To maintain transparency of decisions
• BUT the idea has been taken too far
• Managers and assessors not allowed to talk to
  each other!
• Assessors given too narrow a focus (in terms of
  risk management options, or risk scenario) or too
  broad (model everything because management
  actions not specified)
• No guidance to assessors when data are not
  available
• Managers dont get early warning that r.a.
  request is impossible or useless
• Assessors cant seek guidance, or suggest new
  risk management options when new information
  becomes available
• Managers make assessors follow exposure pathway
  dose-response approach to ensure that a risk
  assessment has been performed

14
Remedies (1) focusing on decisions

• Management must state acceptable level of risk
• All agree zero-risk world not possible, but
  management dont like to be explicit
• Without a threshold risk, any risk can be
  considered unacceptable a greater threat to
  transparency than assessment/management not
  separated
• Focus on the objectives making food safer
• Not performing a risk assessment
• Prioritise by potential benefits of risk
  management actions, not necessarily same as
  prioritising by total health impact
• Consider what is known about the risk problem,
  and data available immediately or within
  acceptable time frame
• Use epidemiological data as much as possible,
  rather than lab data
• Collect more epi data (e.g. Japan, Denmark for
  good examples)
• Consider what analysis could be done with this
  knowledge
• i.e. a risk-based reasoned argument for
  evaluating particular actions
• Critically evaluate assumptions necessary for
  analysing each option
• Look at feasibility, political acceptability and
  compliance issues (risk management overlap)

15
Remedies (2) focusing on decisions

• Estimate the possible pros and cons of a risk
  action
• Secondary risks may outweigh possible benefit
• Recognise that some actions will have impact over
  two or more pathogens
• Note that it may not be possible to evaluate all
  actions
• Perform a cost-benefit analysis on these actions
• Perform a Value of Information analysis
• Determines whether it is worth (in time and
  money) collecting more data before making a
  decision
• Consider strategy to validate whether predicted
  improvement occurs
• And make public intention to review decision if
  prediction incorrect
• Train data producers to supply maximally useful
  data
• E.g. microbiologists taken more than one cfu from
  a plate

16
Draw the problem out
Humans infected with Novovirus
Contaminated food
17
Key component is attribution of risk

• Multiple pathogen sources
• Feedback loops
• Need to use more sophisticated methods for risk
  attribution
• Eg DAGs, classification trees, logistic
  regression
• Need to look at total exposure, eg human-to-human
  exposures and imported food products
• Need to use epidemiological studies, eg
• Casecontrol studies
• Typing of pathogens in food-producing animals and
  humans
• Outbreak investigation
• Comparative risk arguments
• Translate number of reported cases into number of
  illnesses, and then into some consistent measure
  of impact, eg QALY or

18
Example for Salmonella of risk attribution model
19
Prevalence and phage-type distribution of S.
Typhimurium in 1999
20
Consumed amount of relevant food stuffs in
Denmark in 1999
21
The mathematics

• lij Mj pij qi aj, where
• lij the predicted no. of cases/year of type i
  from source j
• Mj the amount of source j available for
  consumption/year
• pij the prevalence of type i in source j
• qi the bacteria-dependent factor for type i
• aj the food-source dependent factor for source j

• qi and aj were included as uniform prior
  distributions

22
Danish Vet Service Salmonella QRAA Bayesian
Approach to Quantify the Contribution of
Animal-food Sources to Human Salmonellosis -
Hald, Vose, Koupeev (2002)
Estimated number of cases of human salmonellosis
in Denmark in 1999 according to source Model
ranks food sources by risk. Updated each year.
Basis of Danish Salmonella program. Now being
applied to antimicrobial resistance.
23
Estimated primary sources of human salmonellosis
in Denmark and year of initiation of control
program
24
Fluoroquinolone-resistant Campylobacter risk
assessment
Model Contaminated carcasses after slaughter
plant probability affected people
25
Broiler
house



Transport

Slaughterhouse
model


Slaughter house

Hanging

Example of Farm-to-Fork model Campylobacter
in poultry Draft report 2001 Institute of Food
Safety and Toxicology Division of Microbiological
Safety Danish Veterinay and Food Administration
Scalding

Defeathering

Evisceration

Washing

Chilling

Export

Chicken parts
Whole
chickens



Chilled
Frozen


Further

Import

Processing

Retail

Catering

Cross

contamination

Consumer
model


Heat

treatment

Consumer



Cross
contamination
Behaves the same way as CVM model if prevalence
is reduced


Heat
treatment
Dose response

Risk estimation

26
In summary

• Risk assessment is a decision tool
• It needs
• Good questions
• Good ideas, creativity
• Rational, decision-focused thinking
• Good use of available data
• Continuous evaluation of fitness-for-purpose
• Continuous communication

27
Thank you Presentation available
atwww.risk-modelling.com