Pharmaceuticals are biologically active compounds that can reach surface waters after use, as conventional wastewater treatment cannot fully remove them. Their stability and specific mode of action may affect non-target aquatic species. However, assessment of their ecotoxicity impacts in Life Cycle Impact Assessment (LCIA) is limited due to the lack of characterization factors (CFs). This study aims to increase the number of pharmaceuticals included in freshwater ecotoxicity impact estimation by deriving CFs. Fate and exposure factors were calculated by extracting physicochemical property data from various sources. Effect factors (EFs) were derived for at least three species groups, based on chronic EC10 values. These EC10s were either directly obtained from empirical data or extrapolated from empirical acute EC50s and chronic NOECs. The 20th percentile of a Species Sensitivity Distribution (HC20) was used as a benchmark to derive EFs. Consequently, CFs were calculated for 271 pharmaceuticals. These newly derived CFs were then compared with those reported in the USEtox model. Furthermore, we compared EFs derived from EC10s extrapolated from acute EC50s versus chronic NOECs for a subset of 59 pharmaceuticals. Finally, we demonstrate the relevance of the newly derived CFs in two LCA applications. Pharmaceuticals with a highly specific mode of action and high toxic potency, such as steroidal hormones (ethinyl estradiol, levonorgestrel, etonogestrel, estradiol, fulvestrant and abiraterone acetate), serotonin modulators (vortioxetine) and pyrethroids (cypermethrin and malathion) exhibit the highest CFs. For pharmaceuticals with specific modes of action, the EFs based on chronic EC10s extrapolated from chronic NOECs can be several orders of magnitude higher than those based on chronic EC10s extrapolated from acute EC50s. This indicates that ecotoxicity may be underestimated for such pharmaceuticals when EC10 values extrapolated from acute EC50s are used. The applications show that the newly derived CFs substantially expand impact coverage and can lead to impact scores that differ by several orders of magnitude from previously published values. This study expands the availability of CFs for pharmaceuticals and improves their robustness by deriving CFs from chronic ecotoxicity data covering at least three species groups. This enables more complete freshwater ecotoxicity assessments in LCIA, supporting decision-making in LCA studies related to the healthcare sector.
Ayeri et al. (Sun,) studied this question.