(218) Assessing Environmental Risks Associated with Contraceptives: A Multi-Scenario Approach

Background: The environmental impact of medicines in surface waters is a growing concern. Traditional assessment methods, relying on periodic and resource-intensive water samplings, underscore the need for efficient estimation methods. While the Predicted Environmental Concentration (PEC) derived through the European Medicines Agency (EMA) approach is widely used, it operates under a worst-scenario assumption. Proposed advancements consider additional determinants, but a unified framework integrating all of them remains elusive.

Objectives: This work aims to establish a framework for factors influencing surface water concentrations of medicines and to use it to provide a multi-scenario environmental risk assessment.

Methods: Framework elements were selected through a literature review, encompassing dispensation, consumption, disposal, metabolism, and wastewater treatment. Scenarios, defined using the contraceptives example, were based on consumption and disposal probabilities. Other elements’ data were obtained from databases and literature. The plausible real-world scenario hypothesized consumption conditioned on administration route, relevant metabolic detoxification, and disposal conditioned on the medicine preparation. The optimal disposal scenario hypothesized absence of disposal in water systems, while the optimal consumption scenario hypothesized the use of all dispensed medicines. Finally, the EMA’s worst-scenario hypothesized no appropriate disposal, no metabolism, and no wastewater treatment. PECs were calculated by adapting the EMA’s formula to each scenario features. Risks were computed as the ratio between the PEC and the Predicted No-Effect Concentration (PNEC) of each scenario.

Results: Dispensations of contraceptives in Italy in 2022 were collected from the Italian Medicines Agency report. When metabolism and wastewater removal data were not available, a 50% default was applied. The first three scenarios indicated a high risk for ethinylestradiol and a moderate risk for levonorgestrel and estradiol. In contrast, the worst-case scenario revealed a high risk for all three molecules. The risk is driven by high consumption for ethinylestradiol, while for estradiol and levonorgestrel, it is influenced by low PNEC and high amount per dose. On the contrary, dienogest, even if highly used, resulted in no risk due to the low quantity per dose and the high PNEC.

Conclusions: Estimating the environmental impact of medicines accounting for their different life-cycle steps and for the features of drug utilization enables a more informed risk assessment compared to the worst-scenario. This framework has the potential to inform decision-making processes, and can be expanded to analyze the environmental impact of other medicines.