Abstract
BackgroundControl of visceral leishmaniasis (VL) on the Indian subcontinent relies on prompt detection and treatment of symptomatic cases. Detection efforts influence the observed VL incidence and how well it reflects the underlying true incidence. As control targets are defined in terms of observed cases, there is an urgent need to understand how changes in detection delay and population coverage of improved detection affect VL control.MethodsUsing a mathematical model for transmission and control of VL, we predict the impact of reduced detection delays and/or increased population coverage of the detection programs on observed and true VL incidence and mortality.ResultsImproved case detection, either by higher coverage or reduced detection delay, causes an initial rise in observed VL incidence before a reduction. Relaxation of improved detection may lead to an apparent temporary (1 year) reduction in VL incidence, but comes with a high risk of resurging infection levels. Duration of symptoms in detected cases shows an unequivocal association with detection effort.ConclusionsVL incidence on its own is not a reliable indicator of the performance of case detection programs. Duration of symptoms in detected cases can be used as an additional marker of the performance of case detection programs.
Highlights
Control of visceral leishmaniasis (VL) on the Indian subcontinent relies on prompt detection and treatment of symptomatic cases
Visceral leishmaniasis (VL), known as kala-azar, is a neglected tropical disease caused by single-celled Leishmania parasites that are transmitted by sandflies [1]
5%–20% of cases develop a skin condition known as post–kala-azar dermal leishmaniasis (PKDL), which lasts several years if left untreated [2]
Summary
Using a mathematical model for transmission and control of VL, we predict the impact of reduced detection delays and/or increased population coverage of the detection programs on observed and true VL incidence and mortality. The competing risk of dying from untreated VL was assumed to increase with duration of symptoms, which was captured using the “linear chain trick” [21] to model progression until death as an Erlang distribution with shape 3. A baseline situation with “standard” detection effort was defined as a situation in which half of the VL cases die undetected, and those who die have symptoms for an average duration of 150 days. These figures are completely unobserved but are consistent with reports of the case ascertainment [22, 23], and were uniquely reproduced by setting the baseline case detection rate to 365/243
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
