FE
DE
RA
L IN
ST
ITU
TE
FO
R R
ISK
AS
SE
SS
ME
NT
Implementing the risk profile:
The German risk assessors`s experience
Dr. Mark Lohmann
Unit Risk Research, Perception, Early
Detection and Impact Assessment
Department Risk Communication
Page 2 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Structure of BfR-Opinions
Page 3 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
53
47
35
15
11
36
42
39
38
38
9
8
18
30
37
2
3
7
16
14
1
2
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
grey box
subject of the assessment
indicated result
indicated reason
opinion as a whole
very understandable
more understandable
some parts
rather difficult to understand
not understandable
53
47
35
15
11
36
42
39
38
38
9
8
18
30
37
2
3
7
16
14
1
2
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
grey box
subject of the assessment
indicated result
indicated reason
opinion as a whole
very understandable
more understandable
some parts
rather difficult to understand
not understandable
Evaluation of BfR-Opinions I
Consumer survey, n = 200, laboratory conditions, duration of the interview: 60 min, evaluation of four opinions, values in %
Page 4 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Effects of risk communication on risk perception and risk understanding of target groups; BfR-Wissenschaft 13/2010
Evaluation of BfR-Opinions II
Consumer survey, n = 200, laboratory conditions, duration of the interview: 60 min, evaluation of four opinions, values in %
78
22
20
20
14
37
22
34
6
28
24
22
2
11
22
15
1
3
12
10
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
I've learned something new
about the topic
I feel well informed after
reading
I have received useful
technical information for
everday life
Reading the opinion helps
me to better deal with the
risk
agree
tend to agree
partly agree
tend to disagree
do not agree
78
22
20
20
14
37
22
34
6
28
24
22
2
11
22
15
1
3
12
10
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
I've learned something new
about the topic
I feel well informed after
reading
I have received useful
technical information for
everday life
Reading the opinion helps
me to better deal with the
risk
agree
tend to agree
partly agree
tend to disagree
do not agree
Page 5 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
On the development of the BfR-Risk Profile
1. Literature research on existing risk profiles
2. Inhouse consultation and evaluation processes:
• Standardised interviews with seven authors of BfR Opinions from five departments
(May-June 2009)
• Experimental online survey among BfR scientists to revise the indicator
“Probability of health impairment” (30 of 243 scientists contacted in writing took
part; February-April 2010)
• Survey of twelve employees in the Risk Communication Department for evaluation
of the revised Risk Profile (November 2010)
• Test phase with the Food Safety Department on the use of the revised Risk Profile
and a new variant; the participants were six authors of BfR Opinions (March-June
2011)
3. External evaluation and feedback processes:
External evaluation (August-October 2011): 47 people from ministries, authorities
and the world of science were contacted in writing; 38 of them returned a
questionnaire.
4. Focus group interviews (August 2015):
Representatives of consumer protection associations, risk management, scientists,
journalists, industry associations, general public
Page 6 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Examples of visualization tools
Product Test
Service 1
Service 2
Service 3
Service 4
Service 5
Cell Phone Service
Page 7 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Numerical assessment of verbal probability information
n = 238; Science writers from USA and Canada
Mosteller and Youtz; Statistical Science 1990, Vol. 5, No. 1, 2-34
Page 8 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Prototypes of the BfR-risk profile for internal
evaluation 1
total population
majority of
population
larger sub-group
smaller
sub-group
single person
not known
certain
probable
possible
improbable
practically
impossible
not known
danger of life
serious impairment
irreversible
serious impairment
reversible
slight impairment
irreversible
slight impairment
reversible
not known
high quality
Fear without
concrete evidence
Generally
recognized proof
plausible suspicion
high quality
not known not known
very great
concern
great concern
moderate concern
low concern
very low
concern
*Estimated from surveys
Distribution
Probability of
Damage
Severity of
Damage
Validity of
Available Data
Trend in Risk
Perception
Distribution Probability of Severity of Validity of *Trend in
Damage Damage Available Data Risk Perception
Page 9 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Prototypes of the BfR-risk profile for
internal evaluation 2+3
Risk Group: Children
spread probability
of damage severity
of damage
validity of
available data
trend in
risk perception
spread
probability of damage
severity of damage
validity of available data
Trend in risk perception
Risk Group: Pregnant
Page 10 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Prototypes of the BfR-risk profile for
internal evaluation 4
Risk Group: Pregnant
spread
probability
of damage
severity
of damage
validity of
available data
trend in
risk perception
Page 11 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Prototype 5: The Slider Model
BfR Risk Profile: Risk Topic X
Distribution not
known
single person
(1)
smaller sub-group
(2)
larger sub-group
majority of the population total
population
Probability of Damage
not known
Practically impossible
Improbable Possible Probable certain
Severity of Damage not
known no health impaiment
slight impairment, reversible
(2)
slight impairment, irreversible
serious impairment, reversible
(1)
serious impairment, irreversible
danger to life
Validity of Available Data
high medium low very low
Controllability by the consumer
control not necessary controllable through
precautionary measures controllable through
avoidance not
controllable
Dark gray fields characterize the properties of the risk assessed in this opinion (as further provided in the text).
Legend:
Single person: < 0,5% of the total population
Smaller sub-group: 0,5% to 25% of the total population)
Certain: probability > 80 % (at least 4 out of 5 cases)
Medium: further research is likely to affect the risk assessment
Reminiscent of changing room symbol
Reflects only infants
Visibility very difficult when presented in small size
Confusion with digestive system
Active seniors do not feel addressed or even feel
offended
Page 12 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Prototype 6 : The Matrix Model
Validity of available data High Medium Low Very Low
Controllability by the
consumer
Control not
necessary
Controllable
through
precautionary
measures
Controllable
through
avoidance
Not
controllable
CertainProbablePossibleImprobablePractically
impossibleNot known
Not known
No
impairment
whole
population
Slight
impairment,
reversible
Slight
impairment,
irreversible
childrenSerious
impairment,
reversible
Serious
impairment,
irreversible
Danger to life
CertainProbablePossibleImprobablePractically
impossibleNot known
Not known
No
impairment
whole
population
Slight
impairment,
reversible
Slight
impairment,
irreversible
childrenSerious
impairment,
reversible
Serious
impairment,
irreversible
Danger to lifeS
eve
rity
of
he
alt
him
pa
irm
en
t
Probability of health impairment
Se
ve
rity
of
he
alt
him
pa
irm
en
t
Probability of health impairment
BfR Risk Profile: Risk TopicBfR Risk Profile: Risk Topic
Page 13 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
The BfR Risk Profile at a glance – example
BfR Risk Profile
Cleaning products with nitric acid concentrations of 20-30 % Opinion no. 041/2010
A Affected group General population
Children
B
Probability of health impairment in the event of contact with cleaning products with a nitric a cid concentration of 20 - 30%
Pra ctically impossible
Improbable Possible
Probable (du e to skin contact or
inhalation of vapours )
Certain ( due to oral
intake )
C
Severity of health impairment in the event of contact with cleaning products with a nitric acid c oncentration of 20 - 30%
No impairment
Slight impairment
M oderate impairment
S erious impairment , reversib le or irreversib l e
D Validity of available data
High: the most important data is avai l able and there are no
contradi c tions
M edium : some important data is missing or
contradictor y
Low : much important data is missing or
contradictory
E C ontroll ability by the consumer [1]
Control not necessary
C ontroll able through precautionary
measures
C ontroll able through avoidance
N ot controllable
Text fields with d ark blue background highlighting characterise the properties of the risk assessed in this Opinion (for more detailed information, please refer to the text in BfR Opinion No. 041/2010 dated 6 September 2010).
It is the aim of the risk profile to visualize the risk described in the opinion. It is not intended to provide risk comparisons. The risk
profile should be read only in conjunction with the opinion.
Row E - Controllability by the consumer
[1] - The indication in the row "Controllability by the consumer" is meant as a descriptive character. BfR assessed household
cleaning products containing 20 to 30% nitric acid as not suitable for use in the household and consequently BfR strongly advise
against the use of these products. BfR recommends further measures to restrict the marketing of such consumer products
containing nitric acid.
Page 14 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Focus group results: benefits and understanding
10
10
10
43
50
33
60
57
57
20
67
50
20
43
10
22
10
17
11
not at all
helpful
little
helpful
partly
helpful helpful
very
helpful
not
specified
Consumer protection
associations/NGOs (n=9)
risk management (n=10)
industry associations (n=10)
scientists (n=7)
journalists (n=9)
general population (n=10)
„understandable“
22 %
33 %
80 %
29 %
30 %
30 %
Page 15 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Focus group results:
Suggestions and wishes for improvement
Traffic light system is preferred for indication
Concrete quantities: What dose, duration, quantity is
dangerous (exposure)?
Concrete recommendations: Should product be consumed or
better not? What are the alternatives?
Demarcation of risky products: Which product groups and / or
producers are affected? (from a specific manufacturer or in
general all products?)
Concrete description of the adverse health effect: What kind
of health impairment is to be expected?
Page 16 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Focus group results: Summary
The majority of the general population considers the BfR risk
profile to be helpful
The majority of experts does not see a concrete use for
the BfR risk profile for their own professional activities.
The BfR risk profile is suitable to decide whether to read
the statements or not
All groups are mainly asking for more specifications
The BfR risk profile has a clearly arranged structure and
gives a first rough impression
Page 17 Mark Lohmann, EFSA-Open Risk Assessment: Methods and Expertise, 16th October 2015, Milan
Acknowledgement
Gaby-Fleur Böl
Astrid Epp
Dept. Risk Communication
Federal Institute for Risk Assesment, Berlin
,BERLIN
FE
DE
RA
L IN
ST
ITU
TE
FO
R R
ISK
AS
SE
SS
ME
NT
Thank you for your attention!
Federal Institute for Risk Assessment
Max-Dohrn-Str. 8-10 10589 Berlin
Tel. +49 30 - 184 12 - 3931 Fax +49 30 - 184 12 - 63931
[email protected] www.bfr.bund.de
Mark Lohmann