Copy the page URI to the clipboard
Webb, Simon Francis
(2001).
DOI: https://doi.org/10.21954/ou.ro.0000fd3a
Abstract
The EU Medicinal Products Directive (65/65/EEC) has been amended to require an environmental risk assessment (ERA) for human pharmaceuticals effective January 1995. At present, official ERA guidelines have yet to be finalised. Previous discussions about their nature have taken place in the absence of a systematic analysis of the potential environmental impacts of pharmaceuticals. This study attempts to address this deficiency via a review of existing ecotoxicity data. Acute ecotoxicity data relating to >100 human pharmaceuticals have been collated. They suggest a lack of acute effects at <100 μg/\ in standard tests. Relative sensitivity based on acute effects was algae (most sensitive) > Daphnia > fish. Chronic effects data were limited and this was identified as a shortcoming. This was reinforced by observations of large differences between acute and chronic responses to steroids in fish. The availability of UK usage data permitted risk characterisation i.e., calculation of PEC/PNEC1 ratios for >60 compounds. Under "worst-case" fate assumptions of no human metabolism, passage of all material to drain, no removal during wastewater treatment and no surface water dilution of effluent, the large majority of pharmaceuticals yielded PEC/PNEC ratios <1 (in theory implying environmental safety). For the remainder, a consideration of surface water dilution and expected wastewater treatment removal was sufficient to yield PEC/PNEC <1. PNEC was based on acute effects data with an assessment factor of 1,000. These assessments ignore the potential for multiple exposure/mixture effects. Calculation of potential lifetime ingestion via drinking water employing "worst-case" assumptions (as above and no removal during drinking water treatment) revealed I70 values (based on ingestion of 2 litres/day for 70 years) generally equivalent to <Z days of the corresponding daily therapeutic doses. Refinement of the exposure calculations or comparisons with monitoring data confirmed the degree of conservatism associated with the "worst-case" exposure estimations.