Collection-informed environmental prioritization of unused household pharmaceuticals using FASS 2025 classification, risk quotients, and TOPSIS-based decision analysis

Unused household pharmaceuticals may contribute to environmental release, yet household accumulation is rarely evaluated alongside available environmental risk information. We integrated collection data from 96 volunteers at three family medicine units in Kütahya, Türkiye, with environmental classifications, source-verified predicted environmental concentration/predicted no-effect concentration (PEC/PNEC)-based risk quotients (RQs), and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Of 12,558 unit doses, 33.89% belonged to six prespecified therapeutic groups selected for environmental relevance. The total Collection-Weighted Environmental Score (CWES) was 919,086.5 mg-score; five active pharmaceutical ingredients (APIs) accounted for 65.39%. Among the eight APIs with source-verified numerical RQs, CWES did not show a clear association with RQ (Spearman’s ρ = −0.167, approximate 95% confidence interval [− 0.782, 0.613], exact two-sided permutation p = 0.703). The reference two-domain TOPSIS analysis retained all 16 APIs using median imputation for eight missing RQs and ranked piracetam, amoxicillin, sertraline, irbesartan, and cephalexin as the highest priorities. Alternative weighting and structural sensitivity analyses changed several individual ranks, whereas the complete-case analysis showed similar ordering among the eight APIs with observed RQs. These rankings may inform further assessment and monitoring but do not represent measured environmental exposure or ecological risk.