Malaria is a life-threatening disease spread to humans through the bites of an infectious female Anopheles mosquito. It is mostly found in tropical countries with prevailing environmental condition of warm humid climate, swampy environments, poor sanitary habits, poverty and ignorance exist.

Globally in 2023, there were an estimated 263 million malaria cases and 597,000 malaria deaths in 83 countries. In 2023, the World Health Organization (WHO) stated that “African Region was home to 94% of malaria cases (246 million) and 95% (569,000) of malaria deaths. Children under 5 accounted for about 76% of all malaria deaths in the Region. Over half of these deaths occurred in four countries: Nigeria (30.9%), the Democratic Republic of the Congo (11.3%), Niger (5.9%) and United Republic of Tanzania (4.3%)” [1].

Compartmental models are often applied to the mathematical modelling of infectious diseases. These models are used to analyze the disease dynamics and to estimate the total number of infected people, the total number of recovered people, and to estimate epidemiological parameters such as the basic reproduction number or effective reproduction number. Such models can show how different public health interventions may affect the outcome of an epidemic. The models are conceptually simple and thus very useful in the field of epidemiology, medicine, economics, genetics, engineering, mathematics etc.

A compartment model partition the population into different compartments (or classes) such as Susceptible (S), Exposed (E), Infected (I), Hospitalized (H), Quarantined (Q), Recovered (R), and Death (D) compartments. The population is assigned to the compartments and people may progress between compartments with change over time t. The acronym for any compartmental model usually show how individuals moves within the compartments; for example, SEIS means susceptible, exposed, infectious, then susceptible again. Compartmental models are most often run with ordinary differential equations (which are deterministic), but can also be used with a stochastic (random) framework, which is more realistic but much more complicated to analyze.

One of the most common compartmental models is the Susceptibility, Infection, and Recovery Models (SIR). The SIR model accurately represents how an infection would spread through a population because it takes into consideration that some people will recover from the disease and no longer be susceptible. Hence, this model assumes that people who recover from the infection become immune and cannot become infected a second time.

Compartmental models may be used to predict properties of how a disease spreads, for example the prevalence (total number of infected) or the duration of an epidemic. In addition, the model allows for understanding how different situations may affect the outcome of the epidemic, e.g., what the most efficient technique is for issuing a limited number of vaccines in a given population.

Throughout recorded history, human society has lived with periodic epidemics and pandemics. Infectious diseases such as bacterial, viral, and parasitic diseases continue to have a significant impact on global health. Numerous disease outbreaks have led to death, societal upheaval, and economic disruption. Hence, a proper understanding of disease dynamics is crucial for monitoring a vaccination program and assessing the risk of future outbreaks.

Nevertheless, in recent years, there has been an increase in research activity regarding compartmental modelling for a complete understanding of disease transmission and persistence. This study seeks to determine the dynamics of spread on the incidence of malaria disease, its infectiousness and control measures using the SEIR-SEI compartmental model.

Mathematical models have a long history of being used to describe the spread of infectious diseases from plague outbreaks more than a century ago to the more recent SARS and Ebola epidemics [2]. From making decisions around different vaccination strategies for influenza to modelling HIV, and from modelling pandemic influenza to currently facilitating real-time policy decision making around the COVID-19 epidemic. However, there is increasing research on mathematical models to fully understand the transmission and persistence of infectious diseases. With ongoing advances in computational tools as well as access to disease incidence data, the use of such models continues to increase.

[3] as well as [4] studied the dynamical behaviour of epidemiological models with nonlinear incidence rates. Their studies show that models with nonlinear incidence have a much wider range of dynamical behaviours than do those with bilinear incidence rates. Again, [5] studied the global dynamics of a SEIR model with varying total population size.

[6] investigated the impact of immigration on the transmission dynamics of tuberculosis. Their theoretical analysis indicated that the disease would persist in the population if there were a prevalence of TB in immigrants. Their study suggests that immigrants have a considerable influence on the overall transmission dynamics behaviour of tuberculosis. Their study show that when there is an immigration of infected people into a population, then the disease-free equilibrium cannot be achieve. [7] incorporated treatment of individuals infectious to TB and chemoprophylaxis (treatment for the latently infected) on the SEIR model. The model assumed that the latently infected individuals develop active disease because of endogenous re-activation, exogenous re-infection and disease relapse. The study shows that treatment of infectious individuals is more effective in the first years of implementation as it cleared active TB immediately. As a result, chemoprophylaxis will do better in controlling the number of infectious due to reduced progression to active TB.

Nevertheless, [8] used the SIR model to describe heterogeneity in vaccination and the cost effectiveness of supplemental immunization activities for measles in Uganda. [9] also developed the SIR model to simulate and better understand the multi-periodic patterns in outbreaks of avian flu in North America. In addition, [10] used a compartmental SEIR model to simulate the dynamics of Ebola outbreak in the Democratic Republic of Congo in 1995. [11] proposed a stochastic discrete time variable of the SIR model for infectious diseases transmission within and between districts, and susceptible individuals interact.

However, Public health organizations throughout the world use such models to evaluate and develop intervention disease outbreak policies for ever-emerging epidemics. Simulation allows for rapid assessment and decision-making, providing quantification and insight into the dynamics of a spread. An intensive inter- and multidisciplinary research effort is speeding up the developments in the field integrating advances from epidemiology, molecular biology, computational engineering and science, and applied mathematics as well as sociology.

To this end, this study seeks to model the dynamics of an infectious disease, using the SEIR-SEI compartmental Model. In addition, to illustrate disease infectiousness to control measures using model on Malaria as case study.

SEIR-SEI compartmental model describes interaction between susceptible, exposed, infected and recovered human population and the susceptible, exposed and infected mosquito population (Figure 2Figure 2). The SEIR-SEI model is evaluated using partial differential equations (which are deterministic), but can also be used with a stochastic (random) framework, which is more realistic but much more complicated to analyze. Individuals in the human population as well as mosquitoes, may move between compartments with change over time t. The SEIR human model part can be used when vital dynamics (births and deaths) are enabled which sustain an epidemic or allow new introductions to spread. Transmission rate depends on the prevalence of infective, the contact rate (biting rate) and the probability of transmission given contact.

Figure 2. Diagram of a SEIR-SEI compartmental model with birth and death (for human and mosquito population).

Table 1 below defines the above model parameters as well as their numerical value and source.

Table 1. Parameter description for SEIR-SEI compartmental model with birth and death (for human and mosquito population).

Hence, the SEIR-SEI model for malaria disease for the population (host and vector) of N h individuals and N m mosquitoes consist of a set of seven differential equations where the solution functions S h , E h , I h and R h for host and S m , E m and I m for the vector are single variable real-valued functions that S h , E h , I h , R h , S m , E m , I m : R 0 + → [ 0 , 1 ] and β h , λ h , σ h , γ h , μ h , β m , λ m , σ m , γ m , μ m ∈ R 0 + .

In addition, the total population proportion at any time t, is assumed to satisfy S h ( t ) + E h ( t ) + I h ( t ) + R h ( t ) = 1 ; and S m ( t ) + E m ( t ) + I m ( t ) = 1

Thus, the SEIR-SEI model for the host-vector population are given as,

where β = τ c ¯ is the effective contact rate, and γ is the removal rate. This means that the expected duration of infection is simply the inverse of the removal rate, i.e. d = γ − 1

In this study, we are more interested in malaria transmission from vector to human. Therefore, from the above differential equations, we would be concerned on the susceptible, exposed infectious and recovered equations of the host population of human. Thus, we can solve the SEIR model part of the human population for a small change in time t, (or short step in time) using the Taylor method, Euler method and the Runge-Kutta method [13]-[15].

Consequently, the Taylor’s method becomes exceedingly complex as its order increases because each increase in order requires a further derivative to be taken for each of our differential equations in the system.

Recall that the Taylor’s expansion of order “n” is given as,

where the dependent variable y i is calculated by short steps of equal time interval h of the independent variable x .

Hence, the Euler method is the Taylor method for order one, i.e.

Thus, the SEIR model with birth and death for the host population with dependent variables S h ( t ) , E h ( t ) , I h ( t ) and R h ( t ) are modeled using the Euler’s iterative method represented by the equations

On the other hand, the Runge-Kutta method is the Taylor method of order four having four sections ( K 1 , K 2 , K 3 , K 4 ) of equations. Thus, for an order four Taylor method, i.e.

the SEIR model with birth and death for the host population with dependent variables S h ( t ) , E h ( t ) , I h ( t ) and R h ( t ) are modeled using the Runge-Kutta’s iterative method represented by the equation

where

Since the SEIR model for host population is a system of four equations, the model will include sixteen separate equations given by,

Thus, in general, Runge-Kutta method of order four SEIR model yields,

Furthermore, from both iterative methods, the results produced by the Runge-Kutta’s method is more accurate than that of the Euler’s method, since it has a forth order accuracy. [13]

The Reproductive Number R is the average number of new infections that one infectious person produces. It measures how many people an infected individual will transmit the disease to before they recover. At the beginning of an epidemic when nearly all are susceptible, it is called R₀. Thus, the reproductive number R refers to the average number of people an infectious person will infect, assuming that the rest of the population is susceptible.

The reproduction number depends on biological, behavioural, social and environmental factors that could foster the spread of the infectious disease. When R > 1 , an epidemic is most likely to occur. But if R < 1 , there could still be (small) cluster of cases, but not enough to sustain an epidemic. In addition, R = 1 is a threshold between epidemic and no epidemic. Note that for a given model, R 0 is fixed over all time.

Figure 3. Graphical depiction of Generations in an epidemic.

This R number is important in determining how quickly an infection will spread through a population. For example, if the value of R for measles is 14, then each case of measles would produce 14 new secondary cases. This would spread through the population much faster than where the value of R was only equal to 2.

Figure 3Figure 3. describes generations in an epidemic, where the number of infections multiplies by a constant factor each day. Here, the reproductive number for the infectiousness above, R = 3 , i.e. an infected person produces three new infections each day and this follows an exponential growth through each generations. This can be used as a simplistic model of how an infection could potentially spread through a population.

However, while the calculation of Reproductive Number R assumes that the entire population is susceptible to the disease, this may not always be the case. Since some people will be immune to the disease due to a prior infection creating life-long immunity, or because of vaccination. Hence, it is imperative to maintain knowledge of the value of Reproductive Number R, in order to control and potentially eradicate a disease from a given population. Thus, at each time period, the effective Reproduction Number R E is calculated as,

where s is the fraction of the host population who are susceptible at time t.

Consequently, our R for SEIR model can be expressed as,

where τ is the transmission probability of infection per contact between a susceptible and infected individual, c ¯ is the average rate of contact between susceptible and infected individuals, and d is the duration of infectious period. ε is the probability of surviving exposure to infection and the inverse of being infected from the exposure, i.e. ε = σ σ − 1

Thus, substituting for the effective contact rate β = τ c ¯ and for the expected duration of infection d = γ − 1 we have,

Consequently, when R < 1 then disease-free equilibrium can be achieved and when it is stable, the number of infected individuals eventually declines to zero, and the disease dies out from the population. On the other hand, the endemic equilibrium is asymptotically stable only when R > 1 , meaning the infection persist in the population because each infected individual infects more than one other person on average.

Herd immunity occurs when a critical proportion of a susceptible population is immunised against a contagious (infectious) disease, giving overall protection to the remainder of the “unprotected” herd (community). Herd immunity works because it is more difficult for diseases to spread between individuals if large numbers are already immune, as this breaks the chain of infectiousness.

Therefore, Herd Immunity Threshold v , is the proportion of the population that needs to be vaccinated or immune in order for an infection to be contained. This is reached when each new case leads to just a single new case, leading to the infection becoming stable within the population. i.e. R E = 1 .

Consequently, at the onset of an infection, everyone in the population is either vaccinated or they are susceptible, so s + v = 1 ; this implies s = 1 − v and substituting this into (35), we have,

Thus, Herd Immunity Threshold

where R is the reproduction number at each period.

The spread of Malaria disease is illustrated below from the given SEIR-SEI model part for the human population. The model employs data gotten from the “World Malaria Report for 2023, by the World Health Organization for Nigeria”. The incidence of Malaria is the estimated number of new and relapse Malaria cases arising in a given year, expressed as the rate per 100,000 population based on WHO World Malaria Report.

The SEIR model for human host, defined the Nigerian Population Size N h as 2279 the total population per 100,000 people. The susceptible group S h is the population of vulnerable people (2185 per 100,000 people) nonimmune to the disease. E h represents the exposed group of individuals, which means the individual has been beaten by mosquitoes but has not been infected malaria disease yet. Thus, the estimated number of individuals exposed to malaria in 2023 is (1741 per 100,000 people). The Infective I h is a group of individuals (681 per 100,000 people) that are sick of malaria disease which are to be treated, vaccinated (or perhaps quarantined or hospitalized so as not to spread the disease until they recover). R h is a group of recovered individuals that come from the infective group, having escaped death through treatment. WHO malaria report estimated 1057 per 100,000 people who recovered from malaria illness in 2023.

The population is open with Nigerian birth rate λ h of 0.035341 and mortality rate μ h of 0.010672 respectively. Nevertheless, WHO duration of illness for an infected person diagnosed with malaria, to recovery may take up to about 2 weeks (14 days) into course of treatment. Thus, the removal rate γ h = 1 14 = 0.071429 . If the transmission rate of contact (biting rate) β h = 0.00035 and the progression rate σ h from exposures to mosquito bites is 0.00031, then, the SEIR model for the spread of malaria disease amongst human population can be modelled as follows:

As at the time of disease incidence (WHO 2023 world malaria report) N h = S h + E h + I h + R h (per 100,000 people). S h ( 0 ) = 2185 , E h ( 0 ) = 1741 , I h ( 0 ) = 681 and R h ( 0 ) = 1057 .

Therefore, we proceed to approximate the rates of change at the initial stage of identification at time t = 0 using the Runge-Kutta’s method.