It's about time: moving away from statistical analysis of ecotoxicity data

Area of Science:

• Environmental science
• Ecotoxicology
• Risk assessment

Background:

• Environmental risk assessment (ERA) traditionally uses single-species lab tests and statistical analyses.
• Current ERA methods focus on toxicity under lab conditions, not real-world impacts.
• Statistical approaches often overlook the critical role of time in chemical toxicity.

Purpose of the Study:

• To argue for the necessity of mechanistic process models over traditional statistical analyses in ERA.
• To highlight the importance of incorporating time-varying exposure in toxicity assessments.
• To advocate for the adoption of toxicokinetic-toxicodynamic (TKTD) models in ERA.

Main Methods:

• Critique of current statistical methods used in ERA.
• Emphasis on the need for mechanistic models that integrate dose and time.
• Discussion of toxicokinetic-toxicodynamic (TKTD) models as a suitable alternative.

Main Results:

• Statistical analyses are insufficient for extrapolating lab toxicity data to real-world scenarios.
• Mechanistic models, specifically TKTD models, are essential for accurate extrapolation.
• Time is a critical, often overlooked, factor in understanding chemical toxicity.

Conclusions:

• ERA should transition from purely statistical analysis to employing mechanistic TKTD models.
• TKTD models provide a more realistic assessment of chemical impacts by considering both dose and time.
• Statistics should be used for error modeling, not for primary process description in ERA.