Local habitual movement as a mechanism for Schistosoma mansoni transmission resurgence: a causal analysis of cross-sectional survey data.

BACKGROUND/AIMS: The WHO aims to eliminate schistosomiasis as a public health problem (EPHP) across 78 endemic countries, with interruption of transmission (IoT) as the goal in selected African countries by 2030. However, for low-prevalence settings that reach EPHP, guidance on managing transmission to maintain EPHP or move towards IoT is limited, partly due to insufficient knowledge of the drivers of persistence and/or re-emergence. In Uganda, some communities inland from Lake Victoria have achieved EPHP for Schistosoma mansoni but not progressed to IoT. This study aimed to assess whether routine, short-range travel to the highly endemic lake could sustain transmission in these settings. METHODS: We conducted a cross-sectional study in five Ugandan villages ~5 km from Lake Victoria. Parasitological data were collected using Kato-Katz and Point-of-Care Circulating Cathodic Antigen tests alongside questionnaires on lake travel from 585 individuals aged 1-91 years. A structural causal model estimated the total and direct effects of travel frequency, activity type, water contact duration and drug treatment history on infection. Bayesian regression models and counterfactual simulations predicted infection under hypothetical interventions. RESULTS: Reaching IoT in low-risk villages may be undermined by habitual short-range travel to high-risk sites driven by the nature and duration of lake contact. Daily lake-related travel caused a 1.7-fold increase in the odds of infection, while occupational activities caused a 3.4-fold increase compared with no lake activity. Counterfactual analysis showed that removing the lake contact duration variable most reduced infection risk among moderate-frequency travellers, while daily travellers showed smaller changes, and some transmission persisted among individuals with little or no lake contact. Simulations demonstrated that modifying lake contact behaviours could reduce individual infection risk but had limited population-level impact. CONCLUSION: These findings indicate that targeting only high-risk villages or individual behaviours is unlikely to achieve sustained, widespread IoT, underscoring the need for integrated control strategies that account for context-specific behaviours and local transmission ecology.