Assessing Strategies Against Gambiense Sleeping Sickness Through Mathematical Modeling

June 1, 2018

Abstract: 

Control of gambiense sleeping sickness relies predominantly on passive and active screening of people, followed by treatment.

Methods

Mathematical modeling explores the potential of 3 complementary interventions in high- and low-transmission settings.

Results

Intervention strategies that included vector control are predicted to halt transmission most quickly. Targeted active screening, with better and more focused coverage, and enhanced passive surveillance, with improved access to diagnosis and treatment, are both estimated to avert many new infections but, when used alone, are unlikely to halt transmission before 2030 in high-risk settings.

Fig 1. Schematic illustrating the Koopman operator for nonlinear dynamical systems.

Schematic of the human African trypanosomiasis transmission cycle, showing baseline medical interventions (A) and complementary interventions using currently available tools considered in this study (B) A, Baseline interventions: passive detection of infected individuals via medical facilities (purple), and active screening (blue). Models with high- and low-risk people assume that high-risk people receive more bites from tsetse (thicker arrow) and only low-risk people are actively screened. B, Additional interventions: (1) Tsetse control (red) directly impacts all transmissions; (2) enhanced passive surveillance improves access and detection at health facilities (purple); (3) targeted active screening improves uptake of active screening campaigns and high-risk people are assumed to participate equally to low-risk people (blue). In some model variants, animals act as a sink to tsetse bites but do not contribute to transmission (dashed arrow).

Conclusions

There was general model consensus in the ranking of the 3 complementary interventions studied, although with discrepancies between the quantitative predictions due to differing epidemiological assumptions within the models. While these predictions provide generic insights into improving control, the most effective strategy in any situation depends on the specific epidemiology in the region and the associated costs.