Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

TREXMO plus: an advanced self-learning model for occupational exposure assessment

Journal of Exposure Science & Environmental Epidemiology volume 30, pages 554–566 (2020)Cite this article

Abstract

In Europe, several occupational exposure models have been developed and are recommended for regulatory exposure assessment. Only some information on the substance of interest (e.g., vapor pressure) and the workplace conditions (e.g., ventilation rate) is required in these models to predict an exposure value that will be later used to characterize the risk. However, it has been shown that models may differ in their predictions and that usually, one of the models best fits a given set of exposure conditions. Unfortunately, there are no clear rules on how to select the best model. In this study, we developed a new modeling approach that together uses the three most popular models, Advanced REACH Tool, Stoffenmanger, and ECETOC TRAv3, to obtain a unique exposure prediction. This approach is an extension of the TREXMO tool, and is called TREXMO+. TREXMO+ applies a machine-learning technique on a set of exposure data with the measured values to split them into smaller subsets, corresponding to exposure conditions sharing similar characteristics. For each subset, TREXMO+ then establishes a regression model with the three REACH tools used as the exposure predictors. The performance of the new model was tested and a comparison was made between the results obtained by TREXMO+ and those obtained by conventional tools. TREXMO+ model was found to be less biased and more accurate than the REACH models. Its prediction differs generally from measurements by a factor of 2–3 from measurements, whereas conventional models were found to differ by a factor 2–14. However, as the available test dataset is limited, its results will need to be confirmed by larger-scale tests.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Schematic illustration of TREXMO+ algorithm and a regression tree example.
Fig. 2: Regression trees established for the evaluated exposure forms.
Fig. 3: Residuals obtained for the training and testing dataset for the three exposure types.
Fig. 4: Measured versus modeled exposure values for three different exposure types.

Similar content being viewed by others

Code availability

To understand how the data was split by using the conditional RT algorithm, the applied R code is given in the Supplementary Information 1.

References

  1. ECHA. Guidance on information requirements and chemical safety assessment—Chapter R.14: occupational exposure assessment. Helsinki (Finland): European Chemical Agency; 2016.

    Google Scholar 

  2. Kromhout H, Symanski E, Rappaport SM. A comprehensive evaluation of within- and between-worker components of occupational exposure to chemical agents. Ann Occup Hyg. 1993;37:253–70.

    CAS  PubMed  Google Scholar 

  3. ECETOC. Addendum to ECETOC targeted risk assessment report no. 93. Brussels: European Centre for Ecotoxicology and Toxicology of Chemicals; 2009.

    Google Scholar 

  4. ECETOC. ECETOC TRA version 3: background and rationale for the improvements—technical report no.114. Brussels: European Centre for Ecotoxicology and Toxicology of Chemicals; 2012.

    Google Scholar 

  5. Kromhout H. Hygiene without numbers. Ann Occup Hyg. 2016;60:403–4.

    Article  Google Scholar 

  6. Kindler P, Winteler R. Anwendbarkeit von Expositionsmodellen für Chemikalien auf Schweizer Verhältnisse. Teilproject 1: Überprüfung der Modelle ‘EASE’ und ‘EMKG-EXPO-TOOL. Zürich (Switzerland): Eidgenössisches Volkwirtschaftsdepartement EDV; Staatsekretariat für Wirtschaft SECO—Arbeitsbedingungen; Chemikalien und Arbeit; 2010.

    Google Scholar 

  7. Fransman W, Van Tongeren M, Cherrie JW, Tischer M, Schneider T, Schinkel J, et al. Advanced Reach Tool (ART): development of the mechanistic model. Ann Occup Hyg. 2011;55:957–79.

    CAS  PubMed  Google Scholar 

  8. Schinkel J, Warren N, Fransman W, van Tongeren M, McDonnell P, Voogd E, et al. Advanced REACH Tool (ART): calibration of the mechanistic model. J Environ Monit. 2011;13:1374–82.

    Article  CAS  Google Scholar 

  9. Marquart H, Heussen H, Le Feber M, Noy D, Tielemans E, Schinkel J, et al. ‘Stoffenmanager’, a web-based control banding tool using an exposure process model. Ann Occup Hyg. 2008;52:429–41.

    PubMed  Google Scholar 

  10. Schinkel J, Fransman W, Heussen H, Kromhout H, Marquart H, Tielemans E. Cross-validation and refinement of the Stoffenmanager as a first tier exposure assessment tool for REACH. Occup Environ Med. 2010;67:125–32.

    Article  Google Scholar 

  11. Tischer M, Lamb J, Hesse S, van Tongeren M. Evaluation of tier one exposure assessment models (ETEAM): project overview and methods. Ann Work Expo Health. 2017;61:911–20.

    Article  Google Scholar 

  12. van Tongeren M, Lamb J, Cherrie JW, MacCalman L, Basinas I, Hesse S. Validation of lower tier exposure tools used for REACH: comparison of tools estimates with available exposure measurements. Ann Work Expo Health. 2017;61:921–38.

    Article  Google Scholar 

  13. Savic N, Gasic B, Schinkel J, Vernez D. Comparing the Advanced REACH Tool’s (ART) estimates with Switzerland’s occupational exposure data. Ann Work Expo Health. 2017;61:954–64.

    Article  CAS  Google Scholar 

  14. Lee EG, Lamb J, Savic N, Basinas I, Gasic B, Jung C, et al. Evaluation of exposure assessment tools under REACH: part I—tier 1 tools. Ann Work Expo Health. 2019;63:218–29.

    Article  Google Scholar 

  15. Lee EG, Lamb J, Savic N, Basinas I, Gasic B, Jung C, et al. Evaluation of exposure assessment tools under REACH: part II—higher tier tools. Ann Work Expo Health. 2019;63:230–41.

    Article  Google Scholar 

  16. Koppisch D, Schinkel J, Gabriel S, Fransman W, Tielemans E. Use of the MEGA exposure database for the validation of the Stoffenmanager model. Ann Occup Hyg. 2011;56:426–39.

    PubMed  Google Scholar 

  17. Kupczewska-Dobecka M, Czerczak S, Jakubowski M. Evaluation of the TRA ECETOC model for inhalation workplace exposure to different organic solvents for selected process categories. Int J Occup Med Environ Health. 2011;24:208–17.

    Article  Google Scholar 

  18. Spinazzè A, Lunghini F, Campagnolo D, Rovelli S, Locatelli M, Cattaneo A, et al. Accuracy evaluation of three modelling tools for occupational exposure assessment. Ann Work Expo Health. 2017;61:284–98.

    Article  Google Scholar 

  19. Savic N, Gasic B, Vernez D. ART, Stoffenmanager, and TRA: a systematic comparison of exposure estimates using the TREXMO translation system. Ann Work Expo Health. 2018;62:72–87.

    Article  Google Scholar 

  20. Savic N, Racordon D, Buchs D, Gasic B, Vernez D. TREXMO: a translation tool to support the use of regulatory occupational exposure models. Ann Occup Hyg. 2016;60:991–1008.

    Article  Google Scholar 

  21. Strasser H, Weber C. On the asymptotic theory of permutation statistics. Mathematical Methods of Statistics. 1999;8;220–50.

  22. Loh W-J. Classification and regression trees. Wiley Interdiscip Rev: Data Min Knowl Discov. 2011;1:14–23.

    Google Scholar 

  23. Garcia CB, Garcia J, Lopez MM, Salmeron R. Collinearity: revisiting the variance inflation factor in ridge regression. J Appl Stat. 2015;42:648–61.

    Article  Google Scholar 

  24. James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning: with application in R. Springer New York:Springer Publishing Company; 2014. ISBN 1461471370.

  25. Cleveland WS. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc. 1979;74:829–36.

    Article  Google Scholar 

  26. R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2018.

    Google Scholar 

  27. Wickham H. Elegant graphics for data analysis. New York: Springer-Verlag; 2016. p. ggplot2.

    Google Scholar 

  28. Riedmann R, Gasic B, Vernez D. Sensitivity analysis, dominant factors, and robustness of the ECETOC TRA v3, Stoffenmanager 4.5, and ART 1.5 occupational exposure models. Risk Anal. 2015;35:211–25.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors are grateful to the Swiss Centre for Applied Human Toxicology (SCAHT) for funding this project and the Swiss Insurance Fund (SUVA) for providing the occupational exposure data.

Funding

The project was funded by the Swiss Centre for Applied Human Toxicology (SCAHT).

Author information

Authors and Affiliations

  1. Center for Primary Care and Public Health (Unisanté), University of Lausanne, Route de la Corniche 2, CH-1066, Epalinges-Lausanne, Switzerland

    Nenad Savic & David Vernez

  2. Health Effects Laboratory Division (HELD), Exposure Assessment Branch (EAB), National Institute for Occupational Safety and Health (NIOSH), 1095 Willowdale Road, Morgantown, WV, 26505, USA

    Eun Gyung Lee

  3. Chemicals and Occupational Health Unit, Swiss State Secretariat for Economic Affairs (SECO), Holzikofenweg 36, CH-3003, Bern, Switzerland

    Bojan Gasic

Authors
  1. Nenad Savic
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Eun Gyung Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Bojan Gasic
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. David Vernez
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Nenad Savic.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Savic, N., Lee, E.G., Gasic, B. et al. TREXMO plus: an advanced self-learning model for occupational exposure assessment. J Expo Sci Environ Epidemiol 30, 554–566 (2020). https://doi.org/10.1038/s41370-020-0203-9

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41370-020-0203-9

Keywords

Search

Advanced search

Quick links