**4. Epidemiologic modelling to study HIV/AIDS dynamics at the macropopulation level**

Computational models and simulations are emerging as vital research tools in the fields of epidemiology, biology, and other sciences. Increasingly, scientific researchers are recognizing the enormous potential of these research tools to solve some of today's biggest and most complex health problems. Computational epidemiology permits the examination and investigation of diseases and risk agents in plants, animals, and humans without jeopardizing lives or creating hazards. This relatively recent branch of science is being used by researchers to understand the overwhelming complexity of the 21st century's health problems. In light of this, computational models that study HIV/AIDS viral dynamics at the macro-population levels by examining the dynamics of HIV/AIDS among different racial groups have been developed by the CCEBRA at Tuskegee University.

Triple Challenges of Psychosocial Factors, Substance Abuse,

educational programs that enhance awareness and counseling.

alternative disease prevention and control strategies.

use status *k*, age *a*, at time *t*, who become AIDS patients at time *t-v*.

(,) (,)

(,) (,) (,)

control strategies.

Mathematically, let:

Similarly:

follows:

drug use status *k*, age *a* at time *t*,

are infected by HIV at time *t-u*.

and HIV/AIDS Risky Behaviors in People Living with HIV/AIDS 141

rates, and parametric variables are very critical in the development of an epidemiologic model. The proportion of people who are not using condoms is the primary focus addressed by this study. Manipulation of the condom use-related variable changes the behavior of the system and results in an increase or decrease of the incidence and prevalence of HIV infections; thus allowing the critical evaluation of alternative HIV/AIDS prevention and

A number of basic assumptions are made in the development of the model. These include the assumptions that: 1) the number of susceptibles at a given time are the total population at that point in time; 2) a susceptible can become infected only if he/she engaged in HIV/AIDS-risky behaviors; 3) an individual can move from high HIV/AIDS-risky behaviors to low HIV/AIDS-risky behaviors or vice versa within the infective subpopulations; and 4) the changes in behavior from high-risk to low-risk may be as a result of

Mathematically, the model parameters are defined as follows. Three major ethnic populations were considered: white, black, and Hispanic, which are designated as ethnic, groups 1, 2, and 3 respectively. Within each group, an individual is considered to be in one of three sex-related statuses: female, heterosexual male, and bisexual/homosexual male, which are designated as sex-related statuses 1, 2, and 3 respectively. Each individual is also considered to be either a non-injecting drug user or injecting drug user. The HIV/AIDS infection rate in a given susceptible population directly depends on the proportion of injecting-drug users, proportion of homosexuals, proportion of people engaged in multiple sexual partnerships, and proportion of people not using condoms. Manipulation of one or several of these variables changes the trend of the system and results in an increase or decrease in the incidence of the HIV/AIDS, thereby supporting the critical evaluation of

*Sijk(a,t)* denote the number of susceptible individuals of ethnic group *i*, sex-related status *j*,

*Iijk(a,t,u)* represent the number of incubating individuals of drug use status *k* (non-injecting drug user or injecting drug user), sex-related status *j*, ethnic group *i*, age *a*, at time *t*, who

*Aijk(a, t, v)* denote the number of AIDS patients of ethnic group *i*, sex-related status *j*, drug

The equations that describe the changes in susceptible populations, HIV-infected populations, and AIDS populations of ethnic group *i*, age *a* at time *t* were then defined as

*S at S at ijk ijk a at S a t ijk ijk ijk t a*

*I at I at I at ijk ijk ijk a t tr u I a t u ijk ijk t au*

{ ( )[1 ( , )] 1} ( , )

{ ( , )[1 ( )] 1} ( , , )

(1)

(2)

#### **4.1 Systems dynamics modelling at the macro-population level**

Systems dynamics (SD) is a concept based on systems thinking where dynamic interaction between the elements of the system is considered in order to study the behavior of the system as a whole. This methodology, introduced in the mid-1950s by Forrester and first described at length in his book *Industrial Dynamics* (1961) with some additional principles presented in his later works (Forrester, 1969; 1971; and 1980), involves development of causal diagrams and computer simulation models that are unique to each problem setting. A central principle of SD is that the complex behaviors of organizational and social systems are the result of ongoing accumulations of people, material or financial assets, information, or even biological or psychological states. Both balancing and reinforcing feedback mechanisms and the concepts of accumulation and feedback have been discussed in various forms for centuries (Richardson, 1991). However, SD uniquely enables the practical application of these concepts in the form of computerized models so that alternative policies and scenarios can be tested in a systematic way that answers the questions of "what if" and "why" (Sterman, 2001).

SD modelling is an iterative process of scope selection, hypothesis generation, causal diagramming, and quantification (Sterman, 2000); it consists of an interlocking set of differential and algebraic equations developed from a broad spectrum of relevant data. A completed SD model may contain scores or even hundreds of equations along with the appropriate numerical inputs. Importantly, epidemiologic SD models are designed to reproduce historical patterns and capable of generating useful insights. The data extrapolated from these epidemiological models are useful not only to study the past, but are reliable also to explore predictive and intervention possibilities (Forrester, 1980; Homer, 1996). With this in mind, a SD model incorporating various HIV/AIDS-risky behaviors has been developed by CCEBRA to model HIV/AIDS.

SD modelling, a tool widely used in epidemiological and mathematical modelling, allows researchers to study and develop a holistic way to assess not only the behavior of the system, but the relationships and interactions between different entities within the system so that scientists can predict what will happen if these systems behaviors persist into the future. If developed carefully, mathematical and statistical models can serve as tools to better understand the epidemiology of HIV/AIDS (Todd et al., 1999). Mathematical models of HIV/AIDS transmission dynamics also play an important role in understanding the epidemiological patterns and methods for disease control as they provide short- and longterm predictions of HIV and AIDS incidence, prevalence, and its dependence on various factors (Todd et al., 1999).

The principles of SD are well suited for modelling and are applicable to HIV/AIDS problems (Dangerfield et al., 2001). The dynamic systems analysis model developed by CCEBRA was performed using the Structural Thinking Experimental Learning Laboratory with Animations (STELLA) software (High Performance Systems, 2000). Applications of systems dynamics methodologies, which employ STELLA software to develop HIV/AIDS models (Dangerfield et al., 2001), addresses the utility of the software in a variety of modelling environments that are suitable for HIV/AIDS modelling purposes.

#### **4.2 The equations that describe the changes in susceptible populations**

The HIV infection rate in a given susceptible population directly depends on the proportion of people engaged in HIV/AIDS-risky behaviors. Equations defining all transition states, rates, and parametric variables are very critical in the development of an epidemiologic model. The proportion of people who are not using condoms is the primary focus addressed by this study. Manipulation of the condom use-related variable changes the behavior of the system and results in an increase or decrease of the incidence and prevalence of HIV infections; thus allowing the critical evaluation of alternative HIV/AIDS prevention and control strategies.

A number of basic assumptions are made in the development of the model. These include the assumptions that: 1) the number of susceptibles at a given time are the total population at that point in time; 2) a susceptible can become infected only if he/she engaged in HIV/AIDS-risky behaviors; 3) an individual can move from high HIV/AIDS-risky behaviors to low HIV/AIDS-risky behaviors or vice versa within the infective subpopulations; and 4) the changes in behavior from high-risk to low-risk may be as a result of educational programs that enhance awareness and counseling.

Mathematically, the model parameters are defined as follows. Three major ethnic populations were considered: white, black, and Hispanic, which are designated as ethnic, groups 1, 2, and 3 respectively. Within each group, an individual is considered to be in one of three sex-related statuses: female, heterosexual male, and bisexual/homosexual male, which are designated as sex-related statuses 1, 2, and 3 respectively. Each individual is also considered to be either a non-injecting drug user or injecting drug user. The HIV/AIDS infection rate in a given susceptible population directly depends on the proportion of injecting-drug users, proportion of homosexuals, proportion of people engaged in multiple sexual partnerships, and proportion of people not using condoms. Manipulation of one or several of these variables changes the trend of the system and results in an increase or decrease in the incidence of the HIV/AIDS, thereby supporting the critical evaluation of alternative disease prevention and control strategies.

Mathematically, let:

*Sijk(a,t)* denote the number of susceptible individuals of ethnic group *i*, sex-related status *j*, drug use status *k*, age *a* at time *t*,

*Iijk(a,t,u)* represent the number of incubating individuals of drug use status *k* (non-injecting drug user or injecting drug user), sex-related status *j*, ethnic group *i*, age *a*, at time *t*, who are infected by HIV at time *t-u*.

#### Similarly:

140 Social and Psychological Aspects of HIV/AIDS and Their Ramifications

Systems dynamics (SD) is a concept based on systems thinking where dynamic interaction between the elements of the system is considered in order to study the behavior of the system as a whole. This methodology, introduced in the mid-1950s by Forrester and first described at length in his book *Industrial Dynamics* (1961) with some additional principles presented in his later works (Forrester, 1969; 1971; and 1980), involves development of causal diagrams and computer simulation models that are unique to each problem setting. A central principle of SD is that the complex behaviors of organizational and social systems are the result of ongoing accumulations of people, material or financial assets, information, or even biological or psychological states. Both balancing and reinforcing feedback mechanisms and the concepts of accumulation and feedback have been discussed in various forms for centuries (Richardson, 1991). However, SD uniquely enables the practical application of these concepts in the form of computerized models so that alternative policies and scenarios can be tested in a systematic way that answers the questions of "what if" and

SD modelling is an iterative process of scope selection, hypothesis generation, causal diagramming, and quantification (Sterman, 2000); it consists of an interlocking set of differential and algebraic equations developed from a broad spectrum of relevant data. A completed SD model may contain scores or even hundreds of equations along with the appropriate numerical inputs. Importantly, epidemiologic SD models are designed to reproduce historical patterns and capable of generating useful insights. The data extrapolated from these epidemiological models are useful not only to study the past, but are reliable also to explore predictive and intervention possibilities (Forrester, 1980; Homer, 1996). With this in mind, a SD model incorporating various HIV/AIDS-risky behaviors has

SD modelling, a tool widely used in epidemiological and mathematical modelling, allows researchers to study and develop a holistic way to assess not only the behavior of the system, but the relationships and interactions between different entities within the system so that scientists can predict what will happen if these systems behaviors persist into the future. If developed carefully, mathematical and statistical models can serve as tools to better understand the epidemiology of HIV/AIDS (Todd et al., 1999). Mathematical models of HIV/AIDS transmission dynamics also play an important role in understanding the epidemiological patterns and methods for disease control as they provide short- and longterm predictions of HIV and AIDS incidence, prevalence, and its dependence on various

The principles of SD are well suited for modelling and are applicable to HIV/AIDS problems (Dangerfield et al., 2001). The dynamic systems analysis model developed by CCEBRA was performed using the Structural Thinking Experimental Learning Laboratory with Animations (STELLA) software (High Performance Systems, 2000). Applications of systems dynamics methodologies, which employ STELLA software to develop HIV/AIDS models (Dangerfield et al., 2001), addresses the utility of the software in a variety of

The HIV infection rate in a given susceptible population directly depends on the proportion of people engaged in HIV/AIDS-risky behaviors. Equations defining all transition states,

modelling environments that are suitable for HIV/AIDS modelling purposes.

**4.2 The equations that describe the changes in susceptible populations** 

**4.1 Systems dynamics modelling at the macro-population level** 

"why" (Sterman, 2001).

factors (Todd et al., 1999).

been developed by CCEBRA to model HIV/AIDS.

*Aijk(a, t, v)* denote the number of AIDS patients of ethnic group *i*, sex-related status *j*, drug use status *k*, age *a*, at time *t*, who become AIDS patients at time *t-v*.

The equations that describe the changes in susceptible populations, HIV-infected populations, and AIDS populations of ethnic group *i*, age *a* at time *t* were then defined as follows:

$$\frac{\partial S\_{\text{ijjk}}(a,t)}{\partial t} + \frac{\partial S\_{\text{ijjk}}(a,t)}{\partial a} = \{\sigma\_{\text{ijjk}}(a)[1 - \gamma\_{\text{ijjk}}(a,t)] - 1\} S\_{\text{ijjk}}(a,t) \tag{1}$$

$$\frac{\partial I\_{\rm ijk}(a,t)}{\partial t} + \frac{\partial I\_{\rm ijk}(a,t)}{\partial a} + \frac{\partial I\_{\rm ijk}(a,t)}{\partial \overline{u}} = \{\sigma\_{\rm ijk}(a,t)[1 - tr(u)] - 1\} I\_{\rm ijk}(a,t,u) \tag{2}$$

Triple Challenges of Psychosocial Factors, Substance Abuse,

preventing disease transmission from infected to healthy individuals.

age *a* at time *t,* who progress to AIDS status during *[t,t+dt)* is represented by:

*<sup>a</sup> ijk <sup>a</sup> da <sup>t</sup> dt ijk <sup>A</sup>*

different ethnic groups are connected through the HIV infection rate,

the model simulations are shown in Figure 1.

the reported findings in the meta-analysis.

**4.3 Simulation results** 

and HIV/AIDS Risky Behaviors in People Living with HIV/AIDS 143

effective in preventing transmission of HIV and other sexually transmitted diseases. Three proportions of condom use at 25%, 50%, and 75% were considered, respectively, in each population to simulate and evaluate the impact on reducing the incidence of HIV by

Once an individual is infected with HIV, after what is most often a long and varied incubation period, the disease advances to the stage of AIDS. The number of individuals of

max

Where *umax* is the maximum incubating time, and *p(u)* is the probability that an individual infected at time *t-u* progresses to AIDS status at time *t*. The systems of equations for

Model parameters were estimated using CDC surveillance data. Computer simulations were carried out on a PowerPC with C as the programming language. In this study, focus was on the use of condoms and its impact on reducing the incidence of the transmission of HIV in sexually active adults in the U.S. The model (Figure 1) shows that if active HIV/AIDS prevention and control interventions are not pursued the HIV/AIDS incidence in the black population would increase from 60 per 100,000 in 1990 to 110 per 100,000 in 2020. In the Hispanic population, incidence would increase from 40 per 100,000 to 68 per 100,000 and in the white population it would increase from around 16 per 100,000 to 23 per 100,000, respectively. This represents an increase in AIDS incidence of 49%, 28%, and 21% for blacks, Hispanics and whites, respectively. As can be seen, these are significant increases for all populations but they are much more devastating for the black subpopulation. Condom use was evaluated in 25% (the status quo used until approximately 1995), 50%, and 75% of sexually active adult populations. The baseline of 25% was used in the model although the rates of condom use varied from low levels of 5% to 10% to 50% or more in previous surveys (CDC, 1996). Figure 1 shows that increased condom use in 50% - 75% of the sexually active population can decrease the rates to the pre-1991 levels, which were 47.9% for blacks, 27.5% for Hispanics, and 11.6% for whites respectively. By the year 2020, the percentage reduction of AIDS would be expected to be 53% in blacks, 49% in Hispanics, and 43% in whites. Previous meta-analysis indicated that condom use could reduce HIV/AIDS by about 69% to 87% (Weller, 1993). A meta-analysis showed that condom use could be effective when used consistently and could potentially reduce HIV by 90% – 95% (Pinkerton and Abramson, 1997). The model simulation examined the proportion of condom use of up to 75%, but if higher levels are evaluated the rate of reduction would be higher and more consistent with

Clearly, the significance of HIV/AIDS and its devastation of minority communities pose a great concern. Among racial/ethnic groups, the impact of HIV is greatest among Blacks. According to the CDC (2008), Blacks, who represent approximately 13% of the U.S. population (U.S. Bureau of Census, 2000), have an estimated rate of HIV diagnoses that is 9 times higher than that of whites and nearly 3 times higher than that of Hispanics. The

*u*

*u*

1 ( , ,0) ( *<sup>a</sup> <sup>t</sup> <sup>u</sup> ijk <sup>p</sup> <sup>u</sup> <sup>I</sup>*

) ( ) ( , , )

*ijk(a,t)*. The results of

(6)

$$\frac{\partial A\_{\rm ijk}(a,t,v)}{\partial t} + \frac{\partial A\_{\rm ijk}(a,t,v)}{\partial a} + \frac{\partial A\_{\rm ijk}(a,t,v)}{\partial v} = \{\mu(v) - 1\} A\_{\rm ijk}(a,t,v) \tag{3}$$

Where the indices *i* = ethnic group status, *j*= sex-related status, and *k* = individuals of drug use status:

*ijk(a)* is the age-specific survival rate of individuals of *i*, *j*, and *k* status;

*ijk(a,t)* is the HIV infection rate of individuals of age *a* at time *t* with *i*, *j*, and *k* status;

tr(u) is the probability that an individual infected by HIV at time *t-u* becomes an AIDS patient at time *t*; and

v) is the survival rate of individuals who become AIDS patients at time *t-v*.

The number of individuals of age *a* at time *t* who acquire HIV by sexual contacts and/or injecting drug use during [*t, t+dt*] is defined as follows:

$$\mathbf{I}\_{\text{ijk}}(\mathbf{a} + \mathbf{d}\mathbf{a}, \mathbf{t} + \mathbf{d}\mathbf{t}, \mathbf{0}) = \sigma\_{\text{ijk}}(\mathbf{a}) \, \gamma\_{\text{jk}}(\mathbf{a}, \mathbf{t}) \, \mathbf{S}\_{\text{jk}}(\mathbf{a}, \mathbf{t}). \tag{4}$$

Note that *Iijk* is the critical HIV infection rate and it relies either on injecting drug use and/or sexual contact. Since interest is to examine condom use as an intervention (HIV preventive approach), how the HIV infection rate via sexual contact is derived is shown.

Let *Fijk(a,t)* denote the events that an individual of age *a*, drug use status *k*, sex related status *j*, in ethnic group *i* is infected by HIV during *[t, t+dt)* due to sexual contact. An individual may have sexual contacts with partners from different ethnic groups. The probability of HIV transmission due to sexual contacts is formulated in terms of the number of partners, number of sexual contacts with each partner, the probability that a partner is infected, and the probability that one contact with an infected partner will result in infection. Since three ethnic groups were considered in this study, each consisting of three sex-related sub groups, the HIV prevalence differs from group to group. The probability that an individual of age *a*, drug use status *k*, sex-related status *j,* in ethnic group *i,* is infected by HIV at time *t*, due to sexual contacts is given by:

$$P\left[F\_{ijk}(a,t)\right] = 1 - \prod\_{e=1}^{\mathfrak{Z}} \left\{1 - q\_{e}(a,t)\right\} \tag{5}$$

Where:

$$q\_{\epsilon}(a,t) = 1 - \left(1 - p\_{jk}(a,t)\left[1 - \left(1 - r\right)^{m\_{\vec{\eta}k,\epsilon}}\right]\right)^{n\_{\vec{\eta}k,\epsilon}}$$


If a condom is used during sexual contact, it is assumed to be 100% protective. Although low levels of condom breaks have been reported condom use is considered to be highly effective in preventing transmission of HIV and other sexually transmitted diseases. Three proportions of condom use at 25%, 50%, and 75% were considered, respectively, in each population to simulate and evaluate the impact on reducing the incidence of HIV by preventing disease transmission from infected to healthy individuals.

Once an individual is infected with HIV, after what is most often a long and varied incubation period, the disease advances to the stage of AIDS. The number of individuals of age *a* at time *t,* who progress to AIDS status during *[t,t+dt)* is represented by:

$$A\_{ijk}(a+da, t+dt, 0) = \sum\_{u=1}^{u\_{\text{max}}} \sigma\_{ijk}(a) p(u) I\_{ijk}(a, t, u) \tag{6}$$

Where *umax* is the maximum incubating time, and *p(u)* is the probability that an individual infected at time *t-u* progresses to AIDS status at time *t*. The systems of equations for different ethnic groups are connected through the HIV infection rate, *ijk(a,t)*. The results of the model simulations are shown in Figure 1.

#### **4.3 Simulation results**

142 Social and Psychological Aspects of HIV/AIDS and Their Ramifications

*A atv A atv A atv ijk ijk ijk v A atv ijk tav*

Where the indices *i* = ethnic group status, *j*= sex-related status, and *k* = individuals of drug

tr(u) is the probability that an individual infected by HIV at time *t-u* becomes an AIDS

The number of individuals of age *a* at time *t* who acquire HIV by sexual contacts and/or

 Iijk(a + da, t + dt, 0) = ijk(a)ijk(a,t) Sijk(a, t). (4) Note that *Iijk* is the critical HIV infection rate and it relies either on injecting drug use and/or sexual contact. Since interest is to examine condom use as an intervention (HIV preventive

Let *Fijk(a,t)* denote the events that an individual of age *a*, drug use status *k*, sex related status *j*, in ethnic group *i* is infected by HIV during *[t, t+dt)* due to sexual contact. An individual may have sexual contacts with partners from different ethnic groups. The probability of HIV transmission due to sexual contacts is formulated in terms of the number of partners, number of sexual contacts with each partner, the probability that a partner is infected, and the probability that one contact with an infected partner will result in infection. Since three ethnic groups were considered in this study, each consisting of three sex-related sub groups, the HIV prevalence differs from group to group. The probability that an individual of age *a*, drug use status *k*, sex-related status *j,* in ethnic group *i,* is infected by HIV at time *t*, due to

<sup>3</sup>

*<sup>n</sup> <sup>m</sup>*

(,) 1 1 (,) 1

*P F at q at ijk <sup>e</sup> e* 

, , ( , ) 1 1 ( , ) 1 (1 ) *ijk e ijk e*

 is the probability that an individual of age *a*, drug use status *k*, sex-related status *j,* in ethnic group *I,* is infected by HIV during *[t, t+dt)* due to sexual contacts with partners

If a condom is used during sexual contact, it is assumed to be 100% protective. Although low levels of condom breaks have been reported condom use is considered to be highly

*<sup>e</sup> je q at p at r*

*r* is the probability of HIV transmission associated with a single sexual contact,

 *mijk,e* is the number of sexual contacts with a partner from ethnic group *e*, and *pje(a,t)* is the probability that a partner from group *e* is infected at time *t*.

*nijk,e* is the number of sexual partners from ethnic group *e*,

*ijk(a,t)* is the HIV infection rate of individuals of age *a* at time *t* with *i*, *j*, and *k* status;

{ ( ) 1} ( , , )

(3)

(5)

(,,) (,,) (,,)

*ijk(a)* is the age-specific survival rate of individuals of *i*, *j*, and *k* status;

v) is the survival rate of individuals who become AIDS patients at time *t-v*.

approach), how the HIV infection rate via sexual contact is derived is shown.

injecting drug use during [*t, t+dt*] is defined as follows:

use status:

patient at time *t*; and

sexual contacts is given by:

from ethnic group *e*,

Where:

Model parameters were estimated using CDC surveillance data. Computer simulations were carried out on a PowerPC with C as the programming language. In this study, focus was on the use of condoms and its impact on reducing the incidence of the transmission of HIV in sexually active adults in the U.S. The model (Figure 1) shows that if active HIV/AIDS prevention and control interventions are not pursued the HIV/AIDS incidence in the black population would increase from 60 per 100,000 in 1990 to 110 per 100,000 in 2020. In the Hispanic population, incidence would increase from 40 per 100,000 to 68 per 100,000 and in the white population it would increase from around 16 per 100,000 to 23 per 100,000, respectively. This represents an increase in AIDS incidence of 49%, 28%, and 21% for blacks, Hispanics and whites, respectively. As can be seen, these are significant increases for all populations but they are much more devastating for the black subpopulation. Condom use was evaluated in 25% (the status quo used until approximately 1995), 50%, and 75% of sexually active adult populations. The baseline of 25% was used in the model although the rates of condom use varied from low levels of 5% to 10% to 50% or more in previous surveys (CDC, 1996). Figure 1 shows that increased condom use in 50% - 75% of the sexually active population can decrease the rates to the pre-1991 levels, which were 47.9% for blacks, 27.5% for Hispanics, and 11.6% for whites respectively. By the year 2020, the percentage reduction of AIDS would be expected to be 53% in blacks, 49% in Hispanics, and 43% in whites. Previous meta-analysis indicated that condom use could reduce HIV/AIDS by about 69% to 87% (Weller, 1993). A meta-analysis showed that condom use could be effective when used consistently and could potentially reduce HIV by 90% – 95% (Pinkerton and Abramson, 1997). The model simulation examined the proportion of condom use of up to 75%, but if higher levels are evaluated the rate of reduction would be higher and more consistent with the reported findings in the meta-analysis.

Clearly, the significance of HIV/AIDS and its devastation of minority communities pose a great concern. Among racial/ethnic groups, the impact of HIV is greatest among Blacks. According to the CDC (2008), Blacks, who represent approximately 13% of the U.S. population (U.S. Bureau of Census, 2000), have an estimated rate of HIV diagnoses that is 9 times higher than that of whites and nearly 3 times higher than that of Hispanics. The

Triple Challenges of Psychosocial Factors, Substance Abuse,

populations and in women (Diaz et al., 1994).

racial and ethnic minority communities.

et al., 1994).

**Materials and methods** 

**Study design** 

respectively, of all PLWHA (Henry J. Kaiser Family Foundation, 2007).

and HIV/AIDS Risky Behaviors in People Living with HIV/AIDS 145

Mississippi, and North Carolina, African Americans accounted for 70%, 84%, and 62%,

Research suggests SES may affect the likelihood of contracting HIV/AIDS (Simon et al., 1995). In a cross-sectional study, Hargreaves (2002) found that men and women of low SES are at greater risk of newly acquired HIV infection. A study of HIV transmission among African-American women in North Carolina found that women with HIV infection were more likely than non-infected women to be unemployed, receive public assistance, have had 20 or more lifetime sexual partners, have used crack or cocaine, or have traded sex for drugs, money, or shelter (CDC, 1982). A lack of SES resources is linked also to HIV/AIDSrisky behaviors and leads to HIV infection (Simon et al., 1995). For each of the HIV/AIDS risk factors examined, low educational level was more common among minority

SES is a key factor that determines the quality of life for PLWHA. Those individuals with fewer resources are often left with limited treatment options (Simon et al., 1995). A study by Diaz and colleagues (1994) indicated that HIV-positive people with lower SES also died sooner than HIV-positive people with higher SES because of their lack of access to medical care, the high cost of antiretroviral drugs, and a lowered immunity from other illnesses. SES is a correlate of behaviors that affect health, access to and use of health care, risk of disease, and mortality (Diaz at al., 1994). When HIV/AIDS rates are examined in light of SES both HIV/AIDS prevalence and incidence are found to be higher among minority populations experiencing high rates of unemployment (Aday, 2001; Fenton, 2001), and lower SES is among the most important determinants of HIV infection among African Americans (Diaz

These findings have significant implications for the development of effective strategies to prevent and treat HIV/AIDS and other health disparities, particularly for the poor and

This study is critical to the development of effective strategies to prevent and control a complex disease challenge such as HIV/AIDS, which is faced by millions of people globally. We conducted a community-based epidemiologic study that integrates multiple determinants – including psychosocial and SES factors – that facilitate HIV/AIDS transmission in all populations. The purpose of this study was to assess the quantitative contributions of each of these factors upon HIV/AIDS transmission. The objectives were: 1) to assess the relationships between psychosocial variables and HIV/AIDS-risky behaviors among PLWHA, and 2) to determine if significant differences exist in substance abuse among PLWHA both before and after their HIV infection status has been established.

The data was collected by a questionnaire instrument survey of the HIV-positive clients of a community-based HIV/AIDS outreach facility (CBHAOF) located in Montgomery, Alabama, USA. The CBHAOF provides treatment and prevention services through education, quality services, and compassionate care for HIV/AIDS clients and their families in 27 counties in Alabama. In addition, the CBHAOF has a medical component/clinic that

**5. Community-based epidemiologic research to address the impact of psychosocial factors and substance abuse on HIV/AIDS-risky behaviors** 

lifetime risk for HIV infection is 1 in 16 for African American men and 1 in 30 for African-American women (Hall, 2008). Hispanics are also disproportionately impacted by HIV/AIDS, representing approximately 15% of the U.S. population but accounting for an estimated 17% of new HIV infections (Hall, 2008). The lifetime risk of an HIV diagnosis is 1 in 36 for Hispanic males and 1 in 106 for Hispanic females (CDC, 2010).

Fig. 1. Projections of AIDS Cases in blacks, Hispanics and whites (under various levels of condom use)

#### **4.3.1 HIV/AIDS as a health disparity in the United States**

Among diseases that disproportionately affect African Americans, the HIV/AIDS epidemic has been particularly devastating for this community at every stage of the disease. Despite extraordinary improvements in HIV treatment, African Americans accounted for 48% of new HIV or AIDS diagnoses in 2005 (CDC, 2007). AIDS remains the leading cause of death among black women between 25-34 years and the second leading cause of death in black men between 35-44 years of age (CDC, 2007). HIV infection levels are especially high (3.6%) among blacks aged 40-49, with males in this age group having an HIV prevalence (4.5%) (McQuillan et al., 2006) that approaches the region-wide prevalence in sub-Saharan Africa (5.0%) (UNAIDS, 2008). The rate of AIDS diagnoses for black adults and adolescents was 10 times the rate for whites and nearly 3 times the rate for Hispanics (CDC, 2007), and for black men it was 8 times more than the rate for white men (CDC, 2007). African-American women had a 23 times greater diagnosis rate than white women (Hader et al., 2001). More than 90 % of babies born with HIV belong to minority groups. African Americans are ten times more likely to die of AIDS than whites (U.S. Department of Health and Human Services, 2008). Studies have also shown higher rates of HIV/AIDS in low-income populations, suggesting

that this pandemic is spreading most rapidly among the poor (Hu et al., 1994). This is significant, especially for the South where half of all African Americans live below 200% of the poverty line. Notably, they have significantly less access to health care than people of other races and ethnicities (Preston et al., 2004; Whetten et al., 2005). In Georgia, Jackson,

lifetime risk for HIV infection is 1 in 16 for African American men and 1 in 30 for African-American women (Hall, 2008). Hispanics are also disproportionately impacted by HIV/AIDS, representing approximately 15% of the U.S. population but accounting for an estimated 17% of new HIV infections (Hall, 2008). The lifetime risk of an HIV diagnosis is 1

Fig. 1. Projections of AIDS Cases in blacks, Hispanics and whites (under various levels of

Among diseases that disproportionately affect African Americans, the HIV/AIDS epidemic has been particularly devastating for this community at every stage of the disease. Despite extraordinary improvements in HIV treatment, African Americans accounted for 48% of new HIV or AIDS diagnoses in 2005 (CDC, 2007). AIDS remains the leading cause of death among black women between 25-34 years and the second leading cause of death in black men between 35-44 years of age (CDC, 2007). HIV infection levels are especially high (3.6%) among blacks aged 40-49, with males in this age group having an HIV prevalence (4.5%) (McQuillan et al., 2006) that approaches the region-wide prevalence in sub-Saharan Africa (5.0%) (UNAIDS, 2008). The rate of AIDS diagnoses for black adults and adolescents was 10 times the rate for whites and nearly 3 times the rate for Hispanics (CDC, 2007), and for black men it was 8 times more than the rate for white men (CDC, 2007). African-American women had a 23 times greater diagnosis rate than white women (Hader et al., 2001). More than 90 % of babies born with HIV belong to minority groups. African Americans are ten times more likely to die of AIDS than whites (U.S. Department of Health and Human Services, 2008). Studies have also shown higher rates of HIV/AIDS in low-income populations, suggesting that this pandemic is spreading most rapidly among the poor (Hu et al., 1994). This is significant, especially for the South where half of all African Americans live below 200% of the poverty line. Notably, they have significantly less access to health care than people of other races and ethnicities (Preston et al., 2004; Whetten et al., 2005). In Georgia, Jackson,

**4.3.1 HIV/AIDS as a health disparity in the United States** 

condom use)

in 36 for Hispanic males and 1 in 106 for Hispanic females (CDC, 2010).

Mississippi, and North Carolina, African Americans accounted for 70%, 84%, and 62%, respectively, of all PLWHA (Henry J. Kaiser Family Foundation, 2007).

Research suggests SES may affect the likelihood of contracting HIV/AIDS (Simon et al., 1995). In a cross-sectional study, Hargreaves (2002) found that men and women of low SES are at greater risk of newly acquired HIV infection. A study of HIV transmission among African-American women in North Carolina found that women with HIV infection were more likely than non-infected women to be unemployed, receive public assistance, have had 20 or more lifetime sexual partners, have used crack or cocaine, or have traded sex for drugs, money, or shelter (CDC, 1982). A lack of SES resources is linked also to HIV/AIDSrisky behaviors and leads to HIV infection (Simon et al., 1995). For each of the HIV/AIDS risk factors examined, low educational level was more common among minority populations and in women (Diaz et al., 1994).

SES is a key factor that determines the quality of life for PLWHA. Those individuals with fewer resources are often left with limited treatment options (Simon et al., 1995). A study by Diaz and colleagues (1994) indicated that HIV-positive people with lower SES also died sooner than HIV-positive people with higher SES because of their lack of access to medical care, the high cost of antiretroviral drugs, and a lowered immunity from other illnesses. SES is a correlate of behaviors that affect health, access to and use of health care, risk of disease, and mortality (Diaz at al., 1994). When HIV/AIDS rates are examined in light of SES both HIV/AIDS prevalence and incidence are found to be higher among minority populations experiencing high rates of unemployment (Aday, 2001; Fenton, 2001), and lower SES is among the most important determinants of HIV infection among African Americans (Diaz et al., 1994).

These findings have significant implications for the development of effective strategies to prevent and treat HIV/AIDS and other health disparities, particularly for the poor and racial and ethnic minority communities.
