ASEI-SEIR model with vaccination for dengue control in Shah Alam, Malaysia

Abstract

Epidemiology modelling provides an understanding of the underlying mechanisms that influence the spread of dengue disease. The most common mathematical models used are the compartment models abbreviated by ASI-SIR, ASEI-SIR and ASEI-SEIR. This paper starts with a discussion of these common models, followed by the derivation of the basic reproduction number (Ro) of each model. The value of Ro in ASI-SIR model is higher than that in ASEI-SIR and ASEI-SEIR models due to the exclusion of exposed adult mosquito in ASI-SIR model. Further, sensitivity analysis on Ro indicates that natural mortality and biting rate of adult mosquito have significant effects on dengue transmission dynamics. Next, an in-house mathematical model named MOSSEIR is developed, based upon the ASEI-SEIR compartment model, in which both mosquito and human populations are considered. The mosquito population is divided into four compartments consisting of aquatic mosquito, susceptible, exposed and infected adult mosquito; while the human population is classified into four compartments comprising susceptible, exposed, infected and recovered human. MOSSEIR is then used to replicate the number of dengue cases in 2010 for Shah Alam, a capital city of Selangor with high incidence of dengue fever. Finally, effectiveness of control strategies, including mosquito breeding sites control, fogging and vaccination, are evaluated for Shah Alam. Simulation results indicate that these three control strategies can significantly reduce dengue transmission, in theory. In reality, the effectiveness of traditional control methods such as elimination of mosquito breeding sites and fogging is below expectation due to non-compliance. Therefore, the adoption of a safe, effective and affordable vaccine remains the best prospect for controlling dengue.