To the Editor.— Lieu et al are to be congratulated for their comprehensive use of decision analysis to model the common problem of the evaluation and treatment of the febrile infant 1 to 3 months of age.1 One of the great strengths of this method is that it can promote discussion and research by making explicit all assumptions about the effectiveness of diagnostic tests and treatments. I would like to thank Dr Lieu for fostering this discussion by making available to me the mathematical representations of this model after its publication.