Numerical analysis for spread of the infection with Japanese encephalitis virus

Abstract

A seroepidemiological survey on hemagglutination-inhibition (HI) antibody titer to Japanese encephalitis (JE) virus in herds of swine in Tokyo metropolitan area during the past 21 years has disclosed a recent trend to dalay in period of antibody acquisition and decline in peak seropositivity rate.Untill 1967 where human infections of JE virus were reported in the capital suburban areas, the HI antibody positivity rate of swine in those areas usually rose sharply late in July to early in August and remained thereafter virtually at a 100% level till October.Over the ensuing years, however, the sharp rise of HI antibody titer became delayed 1-2 months as compared to those before 1968 and the peak seropositivity rate lowered to as low as 20-40%.This paper represents an attempt to numerically analyze the process of spread of an epidemic of JE virus infection with a view to theoretical delineation of the recent changes in mode of JE virus infiltration in herds of swine.We postulated that it might be possible to proceed with the numerical analysis by a model set up on the following presumption. The incidence of swine that may newly contract JE virus infection in an arbitrary unit length of period (Yi) can be expressed in terms of product of three factors;the rate of swine which may serve a potential source of infection in a given swine population (Yi-1), that of susceptible swine in that population {100-(Y0+Y1+…+Yi-1)} and the constant (K) representing the coefficient of swine JE transmission. Thus, we considered the following model for estimation of spread of JE virus infection:Yi=Yi-1 ×{100-(Y0+Yi+…+Yi-1)}× K. The cumulative incidence of swine estimated to newly contract the infection (Yi) during consecutive unit lenghts of period being defined as “infection rate”. this theoretical value was found closely approximately to seropositivity rate determined for HI antibody to JE virus in swine populations studied.Analysis of data with this model has revealed (1) that the rate of swine serving as a potential source of JE virus infection at the ontset of an epidemic of JE has diminished with a consequent delay in sharp rising of HI antibody seropositivity rate, and (2) that the recent decline in transmission coefficient relevant to spread of JE epidemics has led to depression of peak seropositivity rate for HI antibody to JE virus. Thus, these changes in epidemiologic features of outbreaks of JE observed in recent years in the Tokyo metropolitan area are explicable from those which have occurred in respect of the two factors, i. e. decreased subpopulation of swine that would serve as the source of infection and depressed coefficient of transmission of the infection.