Socioeconomic effects in pandemic-induced nicotine use trends revealed by wastewater analysis and machine learning modelling

This research leverages high-resolution wastewater-based epidemiology (WBE) to quantify the impact of the COVID-19 pandemic on community-level nicotine consumption across districts in Türkiye, stratified by socio-economic development. Cotinine, a definitive nicotine metabolite, was analyzed in wastewater from an extensive network of treatment plants to compare pre- and intra-pandemic consumption patterns. Our results reveal an important distinction: nicotine intake markedly decreased in socio-economically in affluent areas, while remaining stable deprived districts. Quantitatively, nicotine consumption remained largely stable in Lower SEDI regions (+1.7 %) but decreased significantly in Middle SEDI (−19.1 %) and Upper SEDI (−22.7 %) regions relative to the pre-pandemic baseline. To model these complex, nonlinear dynamics, an Extreme Learning Machine (ELM) algorithm was employed, demonstrating superior predictive accuracy in capturing the temporal decline in nicotine use linked to pandemic disruptions. This research underscores the critical role of WBE as an efficient and robust engineering tool for real-time public health monitoring. The findings provide actionable insights for policymakers to design targeted, socio-economically sensitive interventions, ultimately supporting the development of more equitable public health strategies during crises.