Use of Hazard Models in Evaluating the Effect of Exposure Duration on the Acute Toxicity of Three Pesticides

Abstract

Renewal toxicity tests (96 h) were conducted with the emulsifiable concentrates of three pesticides: azinphosmethyl, endosulfan, and fenvalerate. A second test design, which incorporated a 6-h period of exposure followed by 90 h of observation, was also conducted on these three pesticides. Toxicity results from the two test types were then compared using different hazard models to test a series of null hypotheses. These comparisons indicated: (1) while the effect of concentration was constant across time, hazard changed for each of the test types; and (2) when the effect of concentration was the same for the initial periods of exposure (6 and 24 h) in the two test designs, the average hazard was significantly lower for 6 h of exposure than for 24 h. This study may have significant implications for relating results from laboratory toxicity tests to mortality among organisms exposed in situ. The 6-h exposure test design provides information on delayed effects due to short-term exposures, while hazard analysis can be used to estimate the underlying time/concentration relationship from standard laboratory toxicity test data.