Skip to Main content Skip to Navigation

Méthodes quantitatives pour évaluer les risques non mutagènes des substances chimiques : Application au cas du chlordécone

Abstract : Risk assessment of chemicals uses for non-mutagenic effects a threshold dose without effect. The aim is to develop a risk assessment framework that allows quantification of non-mutagenic risks. It is inspired by the approach used for genotoxic carcinogens. The added value for safety decision making is shown on the case of chlordecone exposure in Guadeloupe. The official assessment is 1 to 3% of the population are exposed above the threshold (kidney damage). No one has quantified the risks when the threshold is exceeded. However, several million euros (M€) are invested annually for the prevention of exposure to chlordecone in Guadeloupe.The study is based only on published data available in 2013. The analysis of modes of action will identify the possible effects of chlordecone at chronicle low doses. Multidoses experimental studies or epidemiological studies provide data to derive exposure-response functions (ERF). The internal exposure data (blood chlordecone concentration) are available for Guadeloupe before and after the introduction in 2003 of preventive actions. The external exposure data are available only after 2003. Risks, estimated with exposures and ERF, are converted into impacts through the number of people exposed. The impacts are monetized via DALYs and VOLY.Four effects of chlordecone can occur at chronic low dose: prostate and liver cancers, kidney damages and cognitive development impairments. Without threshold and with internal exposure data, the estimated numbers of deaths from liver cancer are respectively before and after 2003: 5.4 /year and 2.0 /year. That is 3.4 deaths avoided per year by reducing exposure. For prostate cancer there are respectively 2.8 and 1.0 /yr (gain: 1.8 /yr) and for kidney damages 0.10 and 0.04 /yr (gain: 0.06 /yr). Loss of IQ points are 1 173/yr and 1 003 /yr (gain = 168 pt./yr). Before 2003, the total annual cost of these impacts is 31.8 million €2006 [10.6 to 64.3]. Impacts on cognitive development account for 62% of the total, liver cancer 27%, prostate cancer 11% and kidney damage 0.6%. After 2003, the cost of impacts is 21.3 M€/yr [5.8 to 42.8]. The benefits due to the reduction of exposures (avoided impacts) are 10.5 M€/yr [4.8 to 21.5]. Compared to annual spending estimated at 3.25 M€/yr, the annual benefit is three times higher. Including a threshold dose in the model do not significantly change the results. Use of the external exposure data, dramatically decreased the costs of impacts down to 4% of the costs estimated with internal exposures data. A sensitivity analysis with Monte Carlo simulation show that the "exposure" variable is the one that most influences the results.This work is the first to consider a quantitative assessment of health risks scenario without threshold, which is more and more often recommended for endocrine disrupters. It is also the first to monetize health impacts of chlordecone in Guadeloupe. This approach proves fruitful of useful information for decision making. It enables to compare basic options such as whether or not a threshold should be used in the model, taking into account any external or internal exposures. It also allows a comparison of the cost of preventive actions with the health benefits they generate. The results obtained facilitate the prioritization of actions for health safety. This quantitative approach is recommended for environmental factors to which are already exposed populations. Actions for reducing exposure to chlordecone in Guadeloupe are justified and efficient in terms of social cost. Epidemiological surveys to study the role of chlordecone in diseases that could not been taken into account for lack of knowledge (adult neurotoxicity, autoimmune diseases, and other developmental effects) would be needed. It would be useful to provide mechanistic studies before epidemiological studies implementation. Extend follow-up of the TIMOUN cohort could reduce uncertainty about cognitive development.
Complete list of metadatas
Contributor : Abes Star :  Contact
Submitted on : Thursday, April 7, 2016 - 5:53:08 PM
Last modification on : Tuesday, June 18, 2019 - 1:34:02 PM
Long-term archiving on: : Monday, November 14, 2016 - 9:51:00 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01299559, version 1



Vincent Nedellec. Méthodes quantitatives pour évaluer les risques non mutagènes des substances chimiques : Application au cas du chlordécone. Médecine humaine et pathologie. Conservatoire national des arts et metiers - CNAM, 2015. Français. ⟨NNT : 2015CNAM0997⟩. ⟨tel-01299559⟩



Record views


Files downloads