Extrapolation factors for small samples of pesticide toxicity data: Special focus on LD50 values for birds and mammals

Authors

  • Robert Luttik,

    Corresponding author
    1. National Institute of Public Health and the Environment, P.O. Box 1, 3720 BA Bilthoven, The Netherlands
    • National Institute of Public Health and the Environment, P.O. Box 1, 3720 BA Bilthoven, The Netherlands
    Search for more papers by this author
  • Tom Aldenberg

    1. National Institute of Public Health and the Environment, P.O. Box 1, 3720 BA Bilthoven, The Netherlands
    Search for more papers by this author

  • Presented in the Symposium on Wildlife Ecotoxicology, 1995 SETAC-Europe Meeting, Copenhagen, Denmark, June 25–28.

Abstract

For the registration procedure of pesticides it is necessary to have specific information and testing data in order to conduct an ecological hazard/risk assessment. The hazard/risk assessment for acute exposure is usually based on a quotient method, where the estimated environmental concentration is compared with the lowest available 50% lethal dose (LD50) value. Generally there are only one or two LD50s available for birds and mammals, and an underestimation of the potential hazard/risk could be a real possibility. In this article, we propose to estimate a hazardous dose for 5% of the species (HD5) for LD50 data, corresponding with the hazardous concentration for 5% of the species for the no-observed effects concentration (NOEC) data. In addition we describe a method for calculating safety factors that can be used in the case of small sample sizes, especially those smaller than 4 and including n = 1. The safety factors to be applied to the geometric mean of the LD50s for a median estimate of the HD5 of birds and mammals are 5.7 and 3.8, respectively. The safety factors for the 95% confidence limit of the HD5 of birds for n = 1, 2, and 3 are 33, 20, and 16, respectively. For mammals these safety factors are 15, 10, and 8.

Ancillary