Related Experiment Video
Updated: May 3, 2026
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
1.7K
Reliability of the Advanced REACH Tool (ART)
Jody Schinkel1, Wouter Fransman, Patricia E McDonnell
11. TNO, PO Box 360, 3700 AJ Zeist, The Netherlands.
The Annals of Occupational Hygiene
|January 23, 2014
Summary
The Advanced REACH Tool (ART) shows improved reliability after training, but significant variability persists among assessors. Further improvements in guidance and training methods are recommended for consistent exposure estimates.
Area of Science:
- Chemical risk assessment
- Occupational hygiene
- Environmental science
Background:
- The Advanced REACH Tool (ART) is utilized for exposure assessment in chemical safety evaluations.
- Assessing the reliability of ART is crucial for its effective implementation in regulatory contexts.
Purpose of the Study:
- To evaluate the inter-assessor agreement for exposure estimates generated by the ART mechanistic model.
- To investigate the impact of training on the reliability of ART assessments.
- To identify factors influencing assessor agreement and estimate reliability.
Main Methods:
- 54 participants underwent a 3-hour training on ART application.
- Assessments were conducted before and after training, compared against gold standard estimates.
- Interclass correlation coefficients (ICCs), percentage agreement, and Cohen kappa statistics were calculated.
Main Results:
- Inter-assessor agreement improved from good to almost perfect after training.
- However, substantial variability remained, with only 42% of post-training estimates within a factor of three of the gold standard.
- Factors like information overload, insufficient guidance, and implementation issues affected reliability.
Conclusions:
- Extensive training is recommended for ART users.
- Improvements in guidance documentation, consensus procedures, and training methods are needed to enhance ART's reliability.
- Considerable variability in exposure estimates between assessors is still expected.
