Modelling Pb blood concentrations in pre-school children living in a residential setting using MERLIN-Expo - Ineris - Institut national de l'environnement industriel et des risques Access content directly
Conference Papers Year : 2017

Modelling Pb blood concentrations in pre-school children living in a residential setting using MERLIN-Expo

Tine Fierens
  • Function : Author
Arnout Standaert
  • Function : Author
Christa Cornelis
  • Function : Author
Philippe Ciffroy
  • Function : Author
Erik Johansson
  • Function : Author
Johan Bierkens
  • Function : Author


We report on model simulations performed using the newly developed exposure tool, MERLIN-Expo, in order to assess lead (Pb) exposure to pre-school children resulting from past emissions by non-ferrous smelters in Belgium (Northern Campine area). Exposure scenarios were constructed to estimate internal Pb exposure in children living in the vicinity of the former industrial sites as compared to children living in adjacent areas and a reference area. Model simulations at individual level are compared with biomonitoring data from a Belgian monitoring campaign (Flemish Government, 2008; Van Holderbeke et al., 2008) in order to verify the model performance of MERLIN-Expo when simulating complex scenarios that account for subject mobility, i.e., duration at the different considered locations (both indoors and outdoors) with varying exposure levels, and individual food consumption patterns. Adopting an assessment approach at individual level improves on more commonly performed generic exposure assessments at population level, by including intake of Pb from local and purchased food products, and taking into account the mobility of the participants, which results in a more comprehensive assessment of individual recent intake of lead. Also, the accuracy of the model predictions as compared to Pb biomonitoring data has been tested via different accuracy approaches and metrics. We calculated a Probability of detection (or true positive rate) of 0.93, reflecting that almost all blood levels above a chosen trigger value of 2.25 mg.L-1 are predicted as such by the model; a Bias of 1.87, i.e. reflecting a slight overestimation of the model, and F1 and F2 scores of 0.65 and 0.8, respectively, confirming that most of the measured positives are detected as such by the model (high F2 score) with a slight overestimation (lower F1 score). Although the model predictions for individual children slightly overpredict the biomonitoring data, they overlap with the prediction interval calculated by MERLIN-Expo based on population averages (Fierens et al., 2016). Modelling exposure at individual level allows risk assessors to attribute the relative contribution of different exposure routes at various locations, reconstructing the exposure history for each individual in a complex site-specific exposure scenario. The slight overprediction of the actual observed Pb blood levels with a factor 2 is compatible with conservative decision-making.


Fichier principal
Vignette du fichier
2017-053_post-print.pdf (217.18 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

ineris-01863856 , version 1 (29-08-2018)


  • HAL Id : ineris-01863856 , version 1


Mirja van Holderbeke, Tine Fierens, Arnout Standaert, Christa Cornelis, Céline Brochot, et al.. Modelling Pb blood concentrations in pre-school children living in a residential setting using MERLIN-Expo. 27. SETAC Europe annual meeting, May 2017, Bruxelles, France. ⟨ineris-01863856⟩


63 View
34 Download


Gmail Mastodon Facebook X LinkedIn More