Here we present and evaluate a framework for estimating concentrations of pharmaceuticals over time in wildlife feeding at wastewater treatment plants (WWTPs). The framework is composed of a series of predictive steps involving the estimation of pharmaceutical concentration in wastewater, accumulation into wildlife food items, uptake by wildlife with subsequent distribution into, and elimination from, tissues. As many pharmacokinetic parameters for wildlife are unavailable for the majority of drugs in use, a read-across approach was employed using either rodent or human data on absorption, distribution, metabolism and excretion (ADME). Comparison of the different steps in the framework, against experimental data for the scenario where birds are feeding on a WWTP contaminated with fluoxetine, showed that: estimated concentrations in wastewater treatment works were lower than measured concentrations; concentrations in food could be reasonably estimated if experimental bioaccumulation data are available; and that read-across from rodent data worked better than human to bird read-across. The framework provides adequate predictions of plasma concentrations and of elimination behavior in birds, but yields poor predictions of distribution in tissues. We believe the approach holds promise, but it is important that we improve our understanding of the physiological similarities and differences between wild birds and domesticated laboratory mammals used in pharmaceutical efficacy/safety trials, so that the wealth of data available can be applied more effectively in ecological risk assessments. This article is protected by copyright. All rights reserved.