Dook Noij

ART in the context of tiered approaches for
Exposure Assessment and Chemical Safety Assessment
Cefic LRI workshop Advanced REACH Tool (ART), 20 April 2012
Dook Noij,
Dow Benelux
Risk assessment under REACH: requirements
To meet the 2010 REACH deadline industry has registered
approximately 1500 classified substances (which require a risk
Assuming on average 10 uses/ES’s per substance and 10
contributing scenarios per ES, this amounts to at least 150,000
assessments in the first registration period
To achieve this, industry needed assessment tools that:
• facilitate mass processing of ES risk assessments with
‘sufficient’ conservatism and ‘acceptable’ uncertainty
• enable a tiered risk assessment approach
• Preferably addressing inhalatory AND dermal exposure
• Conservative in nature
• Screening tool, targeting for use of higher tier tools
• Broad range of application in many type of working environments
• Transparant, relatively easy to use, enable efficient processing
• Clear description of boundaries
Risk of Derm
Tiered approach
Tier 2
Specific models
Tier 1
Tier 1 ½
• More complex models, require expertise in use
• More accurate exposure estimates
• Improved confidence in exposure predictions
• Range of application as broad as possible
(dependent on range of calibration data)
• Clear description of boundaries
Observation: tools used in REACH assessments
Tier 1 tools:
• ECETOC TRA tool v2
• Both for inhalatory and dermal exposure
• Used in approx. 90 – 95 % of the registrations
• Further improved version 3 released this month
Tier 1½ tools:
• Stoffenmanager (inhalatory exposure)
• Risk of Derm (dermal exposure)
Tier 2 tools:
• ART (inhalatory exposure)
• PESTool
Sector Specific Tools :
• ECPA OWB tool
Advanced REACH Tool (ART)
Core: calibrated mechanistic model
Advanced REACH Tool (ART)
Full model including Bayesian module
Position of ART in exposure assessment
• Clearly developed for application in REACH risk assessment
• Contributes to filling the gap for Tier 2 assessment tools
• Application in REACH risk assessments will be focussed on
Tier 2 assessments, where Tier 1 is failing
• May have wider application (however not (yet) ready for
application in CAD exposure assessments)
ART: strengths
• Good conceptual basis (source – receptor model)
• Calibrated with fairly large number of good quality data
• Modifying factor categories and corresponding multipliers
well defined with good justification for numerical values
• Flexibility in setting options for exposure predictions
• Improved confidence in exposure predictions
• Further reduction of uncertainty in the exposure predictions
(provided the added data are representative for the scenario
ART: limitations
• Reliability of predictions in domains where hardly any
measurement data are available for calibration may still be
• Imprecision can be improved by reducing inter-rater
variability; predictions may vary 2-3 orders of magnitude
depending on the rater; this can be improved by means of
guidance, training and experience
• Despite large number of calibration data, accuracy can be
fairly low; larger calibration dataset(s), covering more
sectors and activity types, are needed to improve accuracy
• ART developers can be more explicit on the boundaries of
the tool
Accuracy for liquids and dusts
Variance explained: 66 %
Variance explained: 50 %
2-3 orders of magnitude between estimated (ART) and measured exposure
How to align ART with industry’s needs (1)
Tielemans et al (2011): ‘ART will be further refined in the future and
should therefore be considered an evolving system.’
Changing the model or the calibration factors within the model
during the time period industry is working on REACH assessments
is not acceptable. This would mean that the assessments as
documented in CSR’s, can not be reproduced, when needed.
The only option for doing this, is maintaining several versions of
the ART tool simultaneously
How will ART account for changes in prediction over time due to
updates/additions in the database used as input for the Bayesian
REACH worker risk assessments are aligned to Process Categories
(PROC: ECHA Use Descriptor System); application and
reproducibility of the tool would benefit from development of
prepopulated (default) scenarios for PROC’s
How to align ART with industry’s needs (2)
Application of modifying factors in the assessment implicates
Operational Conditions and Risk Management Measures that not
only have to be documented in the CSR, but also have to be
communicated in the extended SDS. Development of standard
text/phrases that are aligned to the ART modifying factors and can
be applied in SDS systems would greatly support industry’s needs
It will be very difficult to do scaling on exposure scenarios
assessed with ART. Nevertheless it is communicated that scaling
with ART is possible. Assuming this is true, the developers should
clearly explain and document how scaling can be applied and what
the boundaries are of scaling with ART
Enjoy your
ART workshop !!!

similar documents