Upper susceptibility threshold limits with confidence intervals: a method to identify normal and abnormal population...

The interpretation and importance of comparing field values of susceptibility to pesticides with a laboratory reference strain that might bear little resemblance to the actual situation in the field are problematic and a continuing subject of debate. In this paper a procedure for defining a 'normal sensitive' population from a field study of 383 individuals to provide a basis for analysing and interpreting in vitro results is described and examined. Instead of using only the 95th percentile, the upper and lower confidence limits for the 95th percentile were also compared to select the best estimation of the limit for the normal material. A field population constrained by the upper confidence limit for the 95th percentile provides appropriate descriptions of the normal material in this study. This approach should prove useful in studies of pesticide resistance in field populations.