The benefits and drawbacks of using nonparametric methods for estimating product withdrawal times have been debated for many years. This issue was recently revived by Concordet & Toutain (1997a, b) when they described a nonparametric method for withdrawal time estimation. The authors urged the international adoption of this approach, basing their recommendation on three fundamental concerns: (1) the lack of a consistent official procedure for determining a withdrawal time within the European Union (EU); (2) the need to identify a statistical method for improving the international harmonization of withdrawal times for new chemical entities; and (3) a lack of confidence in the robustness of the US Food and Drug Administration/Center for Veterinary Medicine (US FDA) procedure, particularly with respect to minor violations in the underlying parametric assumptions. Due to the critical nature of these issues, the US FDA considers it vital to respond to these concerns. This paper provides a description of the US FDA parametric procedure. We also examine the statistical concerns expressed by Concordet and Toutain, identifying the reasons for our confidence in the US FDA parametric approach. Finally, using their Monte Carlo simulation models, we generate additional datasets to explore the behaviour of their nonparametric procedure and evaluate its ability to support US FDA regulatory activities.

Generalized linear models, Random effects, Model selection, Information criterion,

Our study compares the intelligibility of French-speaking children with a cochlear implant (N = 13) and age-matched children with typical hearing (N = 13) in a narrative task. This contrasts with previous studies in which speech intelligibility of children with cochlear implants is most often tested using repetition or reading tasks. Languages other than English are seldom considered. Their productions were graded by naive and expert listeners. The results show that (1) children with CIs have lower intelligibility, (2) early implantation is a predictor of good intelligibility, and (3) late implantation after two years of age does not prevent the children from eventually reaching good intelligibility.