Table of Contents
ISRN Applied Mathematics
Volume 2012 (2012), Article ID 185939, 14 pages
Research Article

HIV/AIDS Model with Early Detection and Treatment

1General Studies Department, Arusha Technical College, P.O. Box 296, Arusha, Tanzania
2Mathematics Department, University of Dar es Salaam, P.O. Box 35062, Dar es Salaam, Tanzania

Received 21 December 2011; Accepted 23 January 2012

Academic Editors: G. Kyriacou and F. Lamnabhi-Lagarrigue

Copyright © 2012 Augustine S. Mbitila and Jean M. Tchuenche. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


A classical epidemiological framework is used to qualitatively assess the impact of early detection and treatment on the dynamics of HIV/AIDS. Within this theoretical framework, two classes of infected populations: those infected but unaware of their serological status and those who are aware of their disease status, are considered. In this context, we formulate and analyze a deterministic model for the transmission dynamics of HIV/AIDS and assess the potential population-level impact of early detection in curtailing the epidemic. A critical threshold parameter for which case detection will have a positive impact is derived. Model parameters sensitivity analysis indicates that the number of partners is the most sensitive (in increasing the average number of secondary transmission) parameter. However, the case detection coverage is the main drivers in reducing the initial disease transmission. Numerical simulations of the model are provided to support the analytical results. Early detection and treatment alone are insufficient to eliminate the disease, and other control strategies are to be explored.

1. Introduction

HIV/AIDS has killed more than 25 million people globally since its emergence in 1981, making it one of the most destructive epidemics in recorded history. The disease continues to inflict a significant morbidity, mortality, and social-economic and public health burden. For the estimated 33.3 million people living with HIV after nearly 30 years into a very complex epidemic, the gains are real but still fragile, even as the number of annual AIDS-related deaths worldwide has steadily decreased from the peak of 2.1 million in 2004 to an estimated 1.8 million in 2009 [1].

Various preventative and therapeutic measures have been embarked upon, aiming at combating one of the greatest pandemics in modern times [2, 3]. In sub-Saharan Africa, many infected individuals are unaware of their disease status. Recent randomized control trials have found that treating HIV-positive individuals with antiretroviral drugs reduces the risk of them transmitting the disease to their heterosexual partners by more than 90% [4]. As the current treatment therapy has been proven beyond reasonable doubt to reduce transmission, it is imperative to identify those who are infected and put them on treatment when eligible. It is therefore desirable to encourage voluntary testing that will increase case detection, thereby reducing the number of secondary infections of individuals receiving treatment. There is no explicit mathematical account of the potential population level impact of case detection when treatment is available known to us. Thus, a dynamical system model is formulated in order to assess the trade-off/population-level impact between treatment and early detection of HIV positive. The results are sensitive to parameter values, and for this reason, a deterministic sensitivity analysis is carried out.

The rest of this work is organized as follows. The basic model formulation and its analysis are provided in Section 2. The extended model incorporating case detection and treatment is described and analyzed in Section 3. The model simulation using heuristic parameter values for the purpose of illustration follows in Section 4.

2. Model Formulation and Analysis

We begin by formulating a deterministic sex-structured basic HIV/AIDS model (i.e., without interventions). Individuals are identified as male and female only in connection with features peculiar to their sex. The male to female infectivity rate is greater than the female to male [5]. We also assume that the mixing between individuals is homogeneous; individuals may become HIV-infected only through sexual contacts with HIV infected individuals. Those in the final disease stage are considered too ill to remain sexually active. We ignore important HIV transmission path such as intravenous drug injections, vertical transmission, breast feeding, blood transfusion, and needle sharing. It is also assumed that there is no recruitment of HIV positive. The total heterosexual population is divided into male and female subpopulations with the following epidemiological subgroups (the classification is based on individuals disease status): susceptible male and female (𝑆𝑚,𝑆𝑓), infected male and female (𝐼𝑚,𝐼𝑓), and symptomatic individuals in the final disease (AIDS) stage (𝐴𝑚,𝐴𝑓).

New recruits enter the heterosexually active population at constant rates Λ𝑚 and Λ𝑓 for male and female, respectively (all recruits into the population are assumed susceptible). Male and female susceptible acquire infection at time-dependent rates 𝜆𝑚 and 𝜆𝑓 and become infectious. Infectious individuals exhibit AIDS clinical defining symptoms at rates and 𝑧 for male and female, respectively. In the absence of the disease, individuals in the population die of natural death at the rate 𝜇. The disease-induced mortality rate is 𝑑 for both individuals in the infectious and AIDS classes. A full description of the model variables and parameters used in the model is described in Tables 2 and 3, respectively.

Based on our model description and assumptions, we establish the following equations. We note that the red and dash arrows, respectively, in Figures 1 and 2 are the feedback branches which indicate how the male and female subpopulations are coupled (via the force of infection)Male𝑑𝑆𝑚𝑑𝑡=Λ𝑚𝜇+𝜆𝑚𝑆𝑚,𝑑𝐼𝑚𝑑𝑡=𝜆𝑚𝑆𝑚(𝜇+𝑑+)𝐼𝑚,𝑑𝐴𝑚𝑑𝑡=𝐼𝑚(𝜇+𝑑)𝐴𝑚,Female𝑑𝑆𝑓𝑑𝑡=Λ𝑓𝜇+𝜆𝑓𝑆𝑓,𝑑𝐼𝑓𝑑𝑡=𝜆𝑓𝑆𝑓(𝜇+𝑑+𝑧)𝐼𝑓,𝑑𝐴𝑓𝑑𝑡=𝑧𝐼𝑓(𝜇+𝑑)𝐴𝑓.(2.1) The forces of infection for male and female are, respectively, given by𝜆𝑚=𝜂𝑓𝛽𝑚𝐼𝑓𝑁𝑓,𝜆𝑓=𝜂𝑚𝛽𝑓𝐼𝑚𝑁𝑚,(2.2) where 𝛽𝑚, 𝛽𝑓 and 𝜂𝑚, 𝜂𝑓 are, respectively, the probabilities of acquiring HIV and the average number of male and female sexual partners, respectively. The basic model (2.1) with nonnegative initial conditions is epidemiologically meaningful and mathematically well posed. Thus, system (2.1) is dissipative (i.e., all feasible solutions are uniformly bounded [6, 7]). System (2.1) has a disease-free equilibrium (DFE) given by𝐸0=𝑆0𝑚,𝐼0𝑚,𝐴0𝑚,𝑆0𝑓,𝐼0𝑓,𝐴0𝑓=Λ𝑚𝜇Λ,0,0,𝑓𝜇,0,0.(2.3) Using the next-generation operator method [8], the basic model reproduction number 𝑅0 of model (2.1), defined as the number of secondary infections caused by a typical infected individual introduced into the entire susceptible population during his entire period of infectiousness [9], is given by𝑅0=𝛽𝑓𝜂𝑚(𝛽𝜇+𝑑+)𝑚𝜂𝑓(𝜇+𝑑+𝑧).(2.4) The expression of 𝑅0 is a geometric mean of the average number of secondary male infections produced by one female, and the average number of secondary female infections produced by one male. From Theorem 2 of van den Driessche and Watmough [8], the following result holds.

Figure 1: The basic HIV/AIDS model flowchart.
Figure 2: Flow diagram of the HIV/AIDS model with interventions.

Lemma 2.1. The DFE 𝐸0 of system (2.1) is locally asymptotically stable if 𝑅0<1, and unstable if 𝑅0>1.

If 𝑅0<1, then on average an infected individual produces less than one new infection over its infectious period, and the epidemic cannot grow. That is, a small influx of infected individuals would not generate large outbreaks if 𝑅0<1. Conversely, if 𝑅0>1, then each infected individual produces on average more than one new infection, and the disease can invade the population. However, in order for disease elimination to be independent of the initial sizes of the subpopulations of the model when 𝑅0<1, global stability of 𝐸0 is required.

Lemma 2.2. The 𝐸0 is globally asymptotically stable if 𝑅0<1, and unstable otherwise.

Proof. The proof is based on a comparison theorem [10]. The rate of change of the variables representing the infected components of the system (2.1) can be written as 𝑑𝐼𝑚𝑑𝑡𝑑𝐴𝑚𝑑𝑡𝑑𝐼𝑓𝑑𝑡𝑑𝐴𝑓𝐼𝑑𝑡=(𝐹𝑉)𝑚𝐴𝑚𝐼𝑓𝐴𝑓𝜆𝑚𝑆𝑚0𝜆𝑓𝑆𝑓0,(2.5) where the matrices 𝐹 and 𝑉 are given, respectively, by 𝛽𝐹=00𝑚𝜂𝑓Λ𝑚𝜇𝑁𝑓0𝛽0000𝑓𝜂𝑚Λ𝑓𝜇𝑁𝑚0000000,𝑉=𝜇+𝑑+000𝜇+𝑑0000𝜇+𝑑+𝑧000𝑧𝜇+𝑑.(2.6) Thus, 𝑑𝐼𝑚𝑑𝑡𝑑𝐴𝑚𝑑𝑡𝑑𝐼𝑓𝑑𝑡𝑑𝐴𝑓𝐼𝑑𝑡(𝐹𝑉)𝑚𝐴𝑚𝐼𝑓𝐴𝑓.(2.7) Using the fact that the eigenvalues of the matrix 𝐹𝑉 all have negative real parts, it follows that the linearized differential inequality above is stable whenever 𝑅0<1. Consequently, (𝐼𝑚,𝐴𝑚,𝐼𝑓,𝐴𝑓)(0,0,0,0) as 𝑡. By a comparison Theorem [10], (𝐼𝑚,𝐴𝑚,𝐼𝑓,𝐴𝑓)(0,0,0,0) as 𝑡. Thus, (𝑆𝑚,𝐼𝑚,𝐴𝑚,𝑆𝑓,𝐼𝑓,𝐴𝑓)(Λ𝑚/𝜇,0,0,Λ𝑓/𝜇,0,0) as 𝑡 for 𝑅0<1, and hence, the DFE is globally asymptotically stable if 𝑅0<1.

The above result indicates that HIV could be eliminated from the community if the threshold quantity 𝑅0 can be brought to (and maintained at) a value less than unity. The endemic equilibrium (EE) of (2.1) is given by𝐸=𝑆𝑚,𝐼𝑚,𝐴𝑚,𝑆𝑓,𝐼𝑓,𝐴𝑓,(2.8) where𝑆𝑚=Λ𝑚𝜆𝑚+𝜇,𝐼𝑚=Λ𝑚𝜆𝑚𝜆𝑚+𝜇(𝜇+𝑑+),𝐴𝑚=Λ𝑚𝜆𝑚𝜆𝑚,𝑆+𝜇(𝜇+𝑑+)(𝜇+𝑑)𝑓=Λ𝑓𝜆𝑓+𝜇,𝐼𝑓=Λ𝑓𝜆𝑓𝜆𝑓+𝜇(𝜇+𝑑+𝑧),𝐴𝑓=𝑧Λ𝑓𝜆𝑓𝜆𝑓,+𝜇(𝜇+𝑑+𝑧)(𝜇+𝑑)(2.9) where 𝜆𝑚=𝜂𝑓𝛽𝑚(𝐼𝑓/𝑁𝑓) and 𝜆𝑓=𝜂𝑚𝛽𝑓(𝐼𝑚/𝑁𝑚). Solving for 𝜆𝑚 using the values of 𝐼𝑚, 𝐼𝑓 in (2.9) and the value of 𝜆𝑓, after some lengthy algebraic manipulations, the endemic equilibrium of HIV/AIDS basic model satisfies the following linear equation:𝐴𝜆𝑚𝐵=0,(2.10) where𝐴=𝜇𝑁𝑚𝑁𝑓(𝜇+𝑑+)(𝜇+𝑑+𝑧)+𝑁𝑓𝜂𝑚𝛽𝑓Λ𝑚(𝜇+𝑑+𝑧),𝐵=𝜇2𝑁𝑚𝑁𝑓𝜂(𝜇+𝑑+)(𝜇+𝑑+𝑧)1𝑓𝛽𝑚Λ𝑓𝜂𝑚𝛽𝑓Λ𝑚𝜇2𝑁𝑚𝑁𝑓(𝜇+𝑑+)(𝜇+𝑑+𝑧)=𝜇2𝑁𝑚𝑁𝑓(𝜇+𝑑+)(𝜇+𝑑+𝑧)1𝑅20,(2.11)𝐴>0 while 𝐵<0 provided 𝑅0>1, and consequently, the linear system 𝐴𝜆𝑚𝐵=0 has a unique positive solution 𝜆𝑚=𝐵/𝐴, whenever 𝑅0>1. The components of the endemic equilibrium 𝐸 are then determined by substituting 𝜆𝑚=𝐵/𝐴 into (2.9). Noting that 𝑅0<1 implies 𝐵>0; thus, for 𝑅0<1, the force of infection 𝜆𝑚 at steady state is negative (and biologically meaningless). Hence, the model has no endemic equilibrium in this case. Thus, we have established the following result.

Lemma 2.3. The HIV/AIDS model (2.1) has a unique positive EE 𝐸 whenever 𝑅0>1 and none otherwise.

The uniqueness of 𝐸 and the global stability of the DFE imply that the model does not exhibit the phenomenon of backward or subcritical bifurcation where a locally stable EE coexists with a stable DFE when the reproduction number is less than unity. Because the model parameters are taken from different sources, a deterministic sensitivity analysis is carried out using the approach in [11]. Sensitivity indices (of the reproduction number) which measure initial disease transmission allow us to estimate the relative change in a state variable when a parameter changes. The sensitivity indices of 𝑅0 to the parameters for the HIV/AIDS model are given in Table 1. The negative parameters simply means that an increase in that parameter leads to a decrease in the reproductive number. For instance, a 10% increase of the number of sexual partners will lead to a 5% increase of the value of 𝑅0 (initial disease transmission threshold).

Table 1: Numerical values of sensitivity indices of 𝑅0.
Table 2: Model variables and their description.
Table 3: Parameter definitions and their values.

3. Analysis of the Model with Interventions

The basic model is extended to include screening (detected classes, 𝐷𝑚, 𝐷𝑓) and treatment classes (𝑇𝑚, 𝑇𝑓). It is assumed that the number of contacts made by susceptible individuals under treatment is less than or equal to the number of contacts made with an untreated infective due to behavioral change. This is captured via the parameters 𝑟𝑚 and 𝑟𝑓 which are both less than or equal to unity (0𝑟𝑚, 𝑟𝑓1). Also, treatment reduces infectivity, accounted herein by the parameters 𝑝𝑚, 𝑝𝑓<1. The model compartments and flow are depicted in Figure 2, while the additional variables and parameters for the extended model are described in Tables 2 and 3, respectively.

With the above assumptions and terminology, the model is given by the following system of nonlinear equations: Male𝑑𝑆𝑚𝑑𝑡=Λ𝑚𝜆𝑚𝑆+𝜇𝑚,𝑑𝐼𝑚𝑑𝑡=𝜆𝑚𝑆𝑚𝜇+𝑑++𝜎𝑚𝐼𝑚,𝑑𝐷𝑚𝑑𝑡=𝜎𝑚𝐼𝑚𝜇+𝑑+𝜏𝑚𝐷𝑚,𝑑𝑇𝑚𝑑𝑡=𝜏𝑚𝐷𝑚𝜇+𝑑+𝛼𝑚𝑇𝑚,𝑑𝐴𝑚𝑑𝑡=𝛼𝑚𝑇𝑚+𝐼𝑚(𝜇+𝑑)𝐴𝑚,Female𝑑𝑆𝑓𝑑𝑡=Λ𝑓𝜆𝑓𝑆+𝜇𝑓,𝑑𝐼𝑓𝑑𝑡=𝜆𝑓𝑆𝑓𝜇+𝑑+𝑧+𝜎𝑓𝐼𝑓,𝑑𝐷𝑓𝑑𝑡=𝜎𝑓𝐼𝑓𝜇+𝑑+𝜏𝑓𝐷𝑓,𝑑𝑇𝑓𝑑𝑡=𝜏𝑓𝐷𝑓𝜇+𝑑+𝛼𝑓𝑇𝑓,𝑑𝐴𝑓𝑑𝑡=𝛼𝑓𝑇𝑓+𝑧𝐼𝑓(𝜇+𝑑)𝐴𝑓.(3.1) The force of infections of male and female is, respectively, given by 𝜆𝑚=𝜂𝑓𝛽𝑚𝐼𝑓+𝑟𝑓𝐷𝑓+𝑝𝑓𝑇𝑓𝑁𝑓,𝜆𝑓=𝜂𝑚𝛽𝑓𝐼𝑚+𝑟𝑚𝐷𝑚+𝑝𝑚𝑇𝑚𝑁𝑚.(3.2) The DFE of model system (3.1) denoted by 𝐸00 is given by 𝐸00=Λ𝑚𝜇Λ,0,0,0,0,0,𝑓𝜇,0,0,0,0,0.(3.3) Using the next-generation matrix operator [8], the reproduction number of (3.1) is 𝑅𝑇=𝑅𝑚𝑇𝑅𝑓𝑇,(3.4) where 𝑅𝑚𝑇=𝛽𝑓𝜂𝑚𝜇+𝑑++𝜎𝑚𝑟1+𝑚𝜎𝑚𝜇+𝑑+𝜏𝑚+𝑝𝑚𝜏𝑚𝜎𝑚𝜇+𝑑+𝜏𝑚𝜇+𝑑+𝛼𝑚,𝑅𝑓𝑇=𝛽𝑚𝜂𝑓𝜇+𝑑+𝑧+𝜎𝑓𝑟1+𝑓𝜎𝑓𝜇+𝑑+𝜏𝑓+𝑝𝑓𝜏𝑓𝜎𝑓𝜇+𝑑+𝜏𝑓𝜇+𝑑+𝛼𝑓.(3.5) Thus, using Theorem 2 of van den Driessche and Watmough [8], the following result holds.

Lemma 3.1. The DFE 𝐸00 of model system (3.1) is locally asymptotically stable if 𝑅𝑇<1, and unstable if 𝑅𝑇>1.

𝑅𝑚𝑇 and 𝑅𝑓𝑇 are the reproduction numbers for males and females, respectively (i.e., 𝑅𝑚𝑇 represents the number of females infected by a single male during his entire period of infectiousness in a population where treatment is available). If the interventions are dropped, that is, 𝜎𝑚=0, 𝜎𝑓=0, and 𝜏𝑚=0, 𝜏𝑓=0, then the effective reproduction number 𝑅𝑇 reduces to the basic reproduction number 𝑅0. The endemic equilibrium of (3.1) denoted by 𝐸 is 𝐸=𝑆𝑚,𝐼𝑚,𝐷𝑚,𝑇𝑚,𝐴𝑚,𝑆𝑓,𝐼𝑓,𝐷𝑓,𝑇𝑓,𝐴𝑓,(3.6)𝑆𝑚=Λ𝑚𝜇+𝐴𝐼𝑓,𝑆𝑓=Λ𝑓𝜇+𝐵𝐼𝑚,𝐷𝑚=𝜎𝑚𝜇+𝑑+𝜏𝑚𝐼𝑚,𝑇𝑚=𝜎𝑚𝜏𝑚𝜇+𝑑+𝜏𝑚𝜇+𝑑+𝛼𝑚𝐼𝑚,𝑇𝑓=𝜎𝑓𝜏𝑓𝜇+𝑑+𝜏𝑓𝜇+𝑑+𝛼𝑓𝐼𝑓,𝐼𝑚=𝐴Λ𝑓𝐴𝐵Λ𝑚Λ𝑓𝜇2𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚𝜇+𝑑++𝜎𝑚𝜇𝐴𝐵Λ𝑚+𝐴2𝐵Λ𝑚Λ𝑓,𝐼𝑓=𝐴𝐵Λ𝑚Λ𝑓𝜇2𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚𝜇𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚𝐴+𝐴𝐵Λ𝑚,𝐷𝑓=𝜎𝑓𝜇+𝑑+𝜏𝑓𝐼𝑓,𝐴𝑚=𝛼𝑚𝜎𝑚𝜏𝑚+𝜇+𝑑+𝜏𝑚𝜇+𝑑+𝛼𝑚(𝜇+𝑑)𝜇+𝑑+𝜏𝑚𝜇+𝑑+𝛼𝑚𝐼𝑚,𝐴𝑓=𝛼𝑓𝜎𝑓𝜏𝑓+𝑧𝜇+𝑑+𝜏𝑓𝜇+𝑑+𝛼𝑓(𝜇+𝑑)𝜇+𝑑+𝜏𝑓𝜇+𝑑+𝛼𝑓𝐼𝑓,(3.7) where 𝛽𝐴=𝑚𝜂𝑓𝑁𝑓𝑟1+𝑓𝜎𝑓𝜇+𝑑+𝜏𝑓+𝑝𝑓𝜏𝑓𝜎𝑓𝜇+𝑑+𝜏𝑓𝜇+𝑑+𝛼𝑓,𝛽𝐵=𝑓𝜂𝑚𝑁𝑚𝑟1+𝑚𝜎𝑚𝜇+𝑑+𝜏𝑚+𝑝𝑚𝜏𝑚𝜎𝑚𝜇+𝑑+𝜏𝑚𝜇+𝑑+𝛼𝑚.(3.8) The endemic equilibrium exists provided 𝐼𝑚>0 and 𝐼𝑓>0. Consider 𝐼𝑓 given in (3.7), then 𝐼𝑓=𝐴𝐵Λ𝑚Λ𝑓𝜇2𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚𝜇𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚𝐴+𝐴𝐵Λ𝑚=𝐴𝐵Λ𝑚Λ𝑓𝜇2𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚=𝐴𝐵Λ𝑚Λ𝑓𝜇2𝜇+𝑑+𝑧+𝜎𝑓𝜇+𝑑++𝜎𝑚.1(3.9) Substituting 𝐴 and 𝐵 in (3.9), after some rearrangement, we obtain 𝐼𝑓=𝑅2𝑇1.(3.10) It is therefore evident that 𝐼𝑓>0 provided 𝑅𝑇>1. A similar expression can be derived for 𝐼𝑚. Thus, we have established the following result.

Lemma 3.2. The model system (3.1) has a unique positive EE whenever 𝑅𝑇>1 and none otherwise.

We analytically investigate the impact of case detection on HIV/AIDS dynamics. By partially differentiating 𝑅𝑚𝑇 with respect to the case detection rate 𝜎𝑚, we obtain 𝜕𝑅𝑚𝑇𝜕𝜎𝑚=𝛽𝑓𝜂𝑚𝜇+𝑑++𝜎𝑚2(Δ1),(3.11) where Δ=((𝜇+𝑑+)/(𝜇+𝑑++𝜏𝑚))(𝑟𝑚+𝑝𝑚𝜏𝑚/(𝜇+𝑑+𝛼𝑚)). For Δ<1, (3.11) is negative and consequently, early detection will always have a positive impact on the dynamics of HIV/AIDS. If Δ=1, then case detection has no impact (this case will only arise if treatment of those eligible does not follow). From epidemiological and demographical standpoint, the threshold parameter Δ1 is to be expected.

The sensitivity indices of 𝑅𝑇 are given in Table 4. Next, we numerically investigate the impact of the number of partners, case detection, and treatment on the disease dynamics. Tanzania started HIV/AIDS care and treatment in October 2004, and the target for the first year was to cover 44,000 patients. About 96 care and treatment providing facilities were selected to initiate the services, which included four referral hospitals; Muhimbili, Kilimanjaro Christian Medical Centre (KCMC), Bugando Medical Centre (BMC), and Mbeya Medical Centre. Some of the parameter values (see Table 3) are provided courtesy of the regional medical officer estimated based on data from the Bugando Medical Centre (BMC) and the Sekou Toure Hospital (STH) both located in Mwanza City in northern Tanzania. Others are taken from the literature, and the remaining ones are assumed within realistic range for the purpose of illustration.

Table 4: Sensitivity indices of 𝑅𝑇.

Multiple partnerships increase the risk factor of acquiring HIV. When the number of partners is small over a long time period, the rate of infection is minimal. In this case, the disease tends to die down (Figure 3(a)). The disease will persist when multiple and concurrent partnerships are frequent in the community, in which case infections are on the increase over time (Figure 3(b)).

Figure 3: Effect of the number of partners on the dynamics of HIV-positive (a) low sexual activity: 𝜂𝑚=2 and 𝜂𝑓=3, (b) high sexual activity: 𝜂𝑚=4 and 𝜂𝑓=6.

Since it is assumed that all detected HIV-positive individuals are treated if eligible, the shapes of the time trends of detected (Figure 4(a)) and treated individuals (Figure 4(b)) are similar. The slight difference is due to the rate of developing full blown AIDS from the treated class.

Figure 4: Time series of susceptible, detected, and treated individuals.

Figure 5 depicts the effect of increasing early detection and treatment on the reproduction numbers 𝑅0 and 𝑅𝑇.

Figure 5: Graphical representation of 𝑅0 as detection and treatment rates are varied.

4. Conclusion

The dynamic and determinants of the HIV epidemics are multiple and are shaped by the sexual patterns which are related to social, cultural, and economic factors: for example, promiscuity, low and inconsistent condom use, intergenerational sex, concurrent sexual partners, and various opportunistic infections. Screening is a barometer for achieving success in the fight against the epidemic. It is against this background that a simple deterministic HIV/AIDS model which accounts for case detection and antiretroviral therapy was formulated and analyzed. Conditions for the global stability which rules out any possibility for the model to exhibit the phenomenon of backward bifurcation were provided. Therefore, the classical requirement of the reproduction number to be less than unity might be sufficient for disease elimination. However, HIV/AIDS is inherently a multifaceted disease with poverty-drug use-behavioral/attitudinal change and gender inequality are some of the social factors that need to be addressed. Sensitivity results point to the case detection rate as a driving factor in stemming the tide of the epidemic. Thus, increasing voluntary and or mass screening will always have a positive impact on HIV/AIDS control in reducing the disease burden.

Numerical simulations clearly show that early detection and treatment alone are insufficient to eliminate the disease (Figures 4 and 5). Other control strategies such as condom and microbicides used are to be explored. The study is not exhaustive and can be extended in various ways by incorporating a potential (imperfect) vaccine, withdrawal from sexual activity of a fraction of individuals in the AIDS defining stage, and development of drug resistance and superinfection (with different virus strains).


A. S. Mbitila acknowledges with thanks the support in part of the Norad’s programme for Master Studies (in mathematical modelling) at the University of Dar es Salaam and the Arusha Technical College for financial support and study leave.


  1. WHO/UNAIDS, “UNAIDS report on the global AIDS epidemic 2010,” UNAIDS/10.11E JC1958E, 2010. View at Google Scholar
  2. M. A. Foulkes, “Advances in HIV/AIDS statistical methodology over the past decade,” Statistics in Medicine, vol. 17, no. 1, pp. 1–25, 1998. View at Publisher · View at Google Scholar
  3. Z. Mukandavire, J. M. Tchuenche, C. Chiyaka, and G. Musuka, “HIV/AIDS and the use of mathematical models in the theoretical assessment of intervention strategies: a review,” in Advances in Disease Epidemiology, J. M. Tchuenche and Z. Mukandavire, Eds., pp. 221–241, Nova Science Publishers, New York, NY, USA, 2009. View at Google Scholar
  4. “HIV treatment as prevention—it works,” The Lancet, vol. 377, no. 9779, p. 1719, 2011. View at Publisher · View at Google Scholar
  5. N. S. Padian, S. C. Shiboski, and N. P. Jewell, “Female-to-male transmission of human immunodeficiency virus,” Journal of the American Medical Association, vol. 266, no. 12, pp. 1664–1667, 1991. View at Publisher · View at Google Scholar
  6. V. Hutson and K. Schmitt, “Permanence and the dynamics of biological systems,” Mathematical Biosciences, vol. 111, no. 1, pp. 1–71, 1992. View at Publisher · View at Google Scholar
  7. E. Mtisi, H. Rwezaura, and J. M. Tchuenche, “A mathematical analysis of malaria and tuberculosis co-dynamics,” Discrete and Continuous Dynamical Systems Series B, vol. 12, no. 4, pp. 827–864, 2009. View at Publisher · View at Google Scholar
  8. P. van den Driessche and J. Watmough, “Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission,” Mathematical Biosciences, vol. 180, pp. 29–48, 2002. View at Publisher · View at Google Scholar
  9. O. Diekmann, J. A. P. Heesterbeek, and J. A. J. Metz, “On the definition and the computation of the basic reproduction ratio R0 in models for infectious diseases in heterogeneous populations,” Journal of Mathematical Biology, vol. 28, no. 4, pp. 365–382, 1990. View at Publisher · View at Google Scholar
  10. V. Lakshmikantham, S. Leela, and A. A. Martynyuk, Stability Analysis of Nonlinear Systems, Monographs and Textbooks in Pure and Applied Mathematics, Marcel Dekker, New York, NY, USA, 1989.
  11. N. Chitnis, J. M. Hyman, and J. M. Cushing, “Determining important parameters in the spread of malaria through the sensitivity analysis of a mathematical model,” Bulletin of Mathematical Biology, vol. 70, no. 5, pp. 1272–1296, 2008. View at Publisher · View at Google Scholar
  12. Z. Mukandavire and W. Garira, “Sex-structured HIV/AIDS model to analyse the effects of condom use with application to Zimbabwe,” Journal of Mathematical Biology, vol. 54, no. 5, pp. 669–699, 2007. View at Publisher · View at Google Scholar
  13. J. Karrakchou, M. Rachik, and S. Gourari, “Optimal control and infectiology: application to an HIV/AIDS model,” Applied Mathematics and Computation, vol. 177, no. 2, pp. 807–818, 2006. View at Publisher · View at Google Scholar