O-245 Identification of pesticide mixtures to which French agricultural workers are exposed: results from the agriculture and cancer (AGRICAN) cohort study

Introduction

Farmers, particularly in Europe, are exposed to multiple pesticides during their working life. Such exposures can cause adverse health outcomes. The most relevant pesticides mixtures to which French general population is exposed through diet were described last years. We aimed to identify main pesticides mixtures to which French agricultural workers are exposed to, and to classify farmers into clusters based on their mixture exposure profile.

Material and Methods

AGRICAN cohort includes agricultural workers enrolled by questionnaire in 2005–2007, with information on exact year of beginning and end of pesticide use on 13 crops and 5 livestock. We estimated exposure duration to 391 pesticides identified using PESTIMAT crop-exposure matrix for 16,905 men, from 1950 to 2009. We used a Sparse Non-negative Matrix Under-approximation to identify main molecules mixtures based on exposure duration, then, a hierarchical agglomerative clustering to classify farmers sharing similar co-exposure profiles to identified mixtures.

Results

Criterion of relevance and quality of interpretation suggested optimal number of mixtures: from 4 to 45 explaining from 27 to 93% of total variance. We selected 25 mixtures, the best compromise with an explained variance of 83%. Most of identified mixtures contained between 6 to 20 substances. Main mixtures were mostly used on wheat/barley, vine and fruit growing, the two majors mixtures were largely composed of phenoxy herbicides and organophosphorus. Hierarchical clustering method suggested 3 clusters composed of 3,261 to 8,363 men with maximum part of exposure to mixture for cluster 3 to the 1st mixture (11%).

Conclusions

To our knowledge, this is the first study identifying pesticides mixtures in famers and classifying them into clusters based on their mixture exposure profiles. In a second step, we will aim to describe characteristics of the clusters and to study the associations between pesticides mixtures and health outcomes as prostate cancer within AGRICAN cohort.