Combatting Malaysia’s dengue outbreaks with auto-dissemination mosquito traps : a hybrid stochastic-deterministic SIR model

Wells, Jonathan and Greenhalgh, David and Liang, Yanfeng and Megiddo, Itamar and Nazni, Wasi Ahmad and Guat-Ney, Teoh and Lee, Han Lim (2023) Combatting Malaysia’s dengue outbreaks with auto-dissemination mosquito traps : a hybrid stochastic-deterministic SIR model. Communication in Biomathematical Sciences, 6 (2). pp. 169-188. ISSN 2549-2896 (https://doi.org/10.5614/cbms.2023.6.2.7)

[thumbnail of Wells-etal-CBMS-2023-Combatting-Malaysias-dengue-outbreaks-with-auto-dissemination-mosquito-traps]
Preview
 Text. Filename: Wells-etal-CBMS-2023-Combatting-Malaysias-dengue-outbreaks-with-auto-dissemination-mosquito-traps.pdf
Final Published Version
License: Creative Commons Attribution-NoDerivatives 4.0 logo
Download (1MB)| Preview

Abstract

Classical mosquito control methods (e.g. chemical fogging) struggle to sustain long-term reductions in mosquito populations to combat vector-borne diseases like dengue. The Mosquito Home System (MHS) is an auto-dissemination mosquito trap, that kills mosquito larvae before they hatch into adult mosquitoes. A novel hybrid stochastic-deterministic model is presented, that successfully predicts the effect of deploying MHSs within high-rise flats in Selangor, Malaysia. Stochastic SIR (Susceptible-Infected-Recovered) equations (flats) are paired with an existing deterministic SIR model (wider Kuala Lumpur population). Model predictions provide excellent agreement with data from a 44 week MHS trial within the flats. The stochastic model is validated as a powerful tool for predicting short- and long-term impacts of deploying this style of trap within similar environments. Significant, sustainable reductions in mosquito populations are predicted when the MHS is active: with a mean of 9 (95% Uncertainty Range (UR): 1; 30) during the 44 week trial period, compared to 35 (95% UR: 1; 234) dengue cases with no MHSs. Long-term predictions for endemic equilibrium show MHSs significantly narrow the mosquito population distribution and reduce dengue prevalence: from a mean of 5 (95% UR: 0; 52) (no MHS), to 1 (95% UR: 0; 8) dengue cases annually (with MHS).

CORE (COnnecting REpositories)