One Health interventions for leptospirosis: Do we need an engineer?

Introduction: Leptospirosis is a globally important zoonotic disease that occurs mainly in tropical or sub-tropical countries. Whilst rodents and domestic mammals are considered the major reservoir hosts, there is debate about the role the environment (soil) plays as a reservoir for infection. Fiji, a small island nation in the South Pacific, reports a high burden of severe and often fatal leptospirosis. This study used systems dynamics modelling (SDM) to explore options for a One Health intervention to prevent infection as an adjunct to recently updated clinical guidelines and a renewed focus on early detection of cases. Methods: A multidisciplinary team used a participatory model-building process to develop a qualitative (causal loop diagram (CLD)) representation of the system structures that drive the incidence of leptospirosis in Fiji. The CLD was informed by ecological time-series analysis to determine associations between weather variables and case incidence and an analysis of data from clinical records of leptospirosis cases to determine clinical parameters for disease and its treatment. A quantitative SDM was subsequently developed and used to simulate sectoral and multisectoral interventions to reduce the incidence and mortality associated with leptospirosis. Results: There was a positive association between rainfall and weekly incidence of leptospirosis with a 2-3-week lag and a variable increased risk associated with increased in maximum and minimum air temperature. Data from 578 cases showed that a delay in diagnosis was a critical factor driving mortality, with health seeking occurring 5.0 ± 4.2 days after first symptoms, 20% of patients developing severe leptospirosis and 50 subsequent deaths (58% within 24 hours of admission). The CLD contained three major subsystems: health seeking behaviour and clinical management; water and soil factors (weather, soil and run off); and animal populations. Simulated interventions demonstrated the importance of managing overland water flow and improving drainage infrastructure as critical leverage points for reducing infection risk. Interventions based on surveillance in animals and health interventions alone were ineffective. Discussion: Leptospirosis occurs in a complex interconnected system with many variables outside human and animal health systems. The SDM realistically incorporates a diverse range of environmental, behavioural and infrastructure variables to demonstrate the underlying system structures and drivers of leptospirosis incidence and burden. The importance of environmental reservoirs such as soil support a shift in mental models away from a focus on the abundance and infection status of animal reservoirs, which are only required periodically to “top-up” the reservoir. The SDM allows us to simulate more complex interventions that include infrastructure management and hydrology. Conclusion: The results of this study show that interventions for diseases with strong ecological drivers such as leptospirosis need One Health innovation that considers the hydrology and roles of engineering to reduce underlying drivers.