A model for predicting the frequency of high pesticide exposure events in the agricultural health study

The frequency of self-reported high pesticide exposure events (HPEE) has be en recorded in the NCI/EPA/NIEHS Agricultural Health Study (AHS). Fourteen percent (14%) of the enrolled applicators responding reported "an incident or experience while using any pesticide which caused an unusually high expo sure." These data show, as expected, that the probability of a report of an HPEE increases with the cumulative number of days of pesticide application reported by the applicator. We have developed a three-parameter model that predicts the risk odds ratio (OR) of an HPEE as a function of the number o f days that pesticides are applied. These events are costly in terms of res ulting health-care visits, lost time from work, and potential risk for canc er and other chronic diseases. We propose that failure to carefully follow all the pesticide manufacturer's label requirements, inexperience, and rand om events (i.e., breaking hose) are the three factors responsible for the e vents observed. Special precautions for new or infrequent users of pesticid es are indicated. (C) 2000 Academic Press.