Exploiting monitoring data in environmental exposure modelling and risk assessment of pharmaceuticals

In order to establish the environmental impact of an active pharmaceutical ingredient (API), good information on the level of exposure in surface waters is needed. Exposure concentrations are typically estimated using information on the usage of an API as well as removal rates in the patient, the wastewater system and in surface waters. These input data are often highly variable and difficult to obtain, so model estimates often do not agree with measurements made in the field. In this paper we present an approach which uses inverse modelling to estimate overall removal rates of pharmaceuticals at the catchment scale using a hydrological model as well as prescription and monitoring data for a few representative sites for a country or region. These overall removal rates are then used to model exposure across the broader landscape. Evaluation of this approach for APIs in surface waters across England and Wales showed good agreement between modelled exposure distributions and available monitoring data. The use of the approach, alongside estimates of predicted no-effect concentrations for the 12 study compounds, to assess risk of the APIs across the UK landscape, indicated that, for most of the compounds, risks to aquatic life were low. However, ibuprofen was predicted to pose an unacceptable risk in 49.5% of the river reaches studied. For diclofenac, predicted exposure concentrations were also compared to the Environmental Quality Standard previously proposed by the European Commission and 4.5% of river reaches were predicted to exceed this concentration. While the current study focused on pharmaceuticals, the approach could also be valuable in assessing the risks of other ‘down the drain’ chemicals and could help inform our understanding of the important dissipation processes for pharmaceuticals in the pathway from the patient to ecological receptors. •Model parameterisation with drug monitoring data improves exposure predictions.•We don't fully understand environmental dissipation processes of pharmaceuticals.•A significant percentage of river reaches appear at risk from pharmaceutical exposure.