Publication Type: Conference Paper/Unpublished Manuscript
Abstract: In 2005, about 5.5 million people in South Africa were estimated to be living with HIV. Although it appears that the increase in prevalence has slowed and stabilized over the last 5 years, there is no evidence that explains why this may be happening. The objective of this study is to present analyses of two national surveys that show the number of HIV infections averted as of 2005 by means of prevention behaviour and the impact that AIDS communication programs have had on HIV prevention behaviour as of 2006. Cross-sectional, nationally representative household surveys were conducted in 2005 and 2006 to measure HIV prevalence, HIV prevention behaviour, and exposure to AIDS communication programs designed to promote prevention behaviour and help people living with HIV. The result showed that exposure to 19 AIDS communication programs had a substantial, monotonically increasing impact on using condoms to prevent HIV infection and eleven other AIDS-related outcomes. Propensity score analysis revealed that the HIV negative status among those who practiced some kind of prevention behaviour would have been 86.2% rather than 90.5% if they had not practiced a 4.3 percentage point difference, amounting to an estimated 718,197 HIV infections averted by 2005, and a 15.3% reduction in HIV infections for an estimated 20-year lifetime savings of this reduction in HIV infections would be US 5.7 billion dollars.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: The evidence-base for mass media HIV prevention campaigns in the US is weak and the theoretical frameworks for such interventions are not well developed. In this paper we describe an approach to using culturally sensitive mass media messages to change the beliefs and behavior of African-American adolescents at risk for HIV. This approach exploits the potential that mass media messages have, not only to reach a large segment of the adolescent population and thereby supporting normative change, but also to engage the most at-risk segments of this group to change their behavior. We illustrate the effectiveness of this approach with results from an ongoing HIV-prevention trial directed to African-American adolescents in two medium sized cities in the US.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: While the prevalence of HIV in the United States continues to be higher for males than for females, the infection rate of females is growing rapidly. Of vital importance in this trend is the impact of HIV testing in women. The importance of HIV testing for women is two-fold. First, it provides a method for primary prevention of infection since women who test negative can be educated on ways to prevent future transmission. Second, the mortality rates for women who test positive can be greatly reduced with the onset of early treatment. The 2002 cycle of the National Survey of Family Growth (NSFG) includes information about HIV/STD testing including the source of the HIV test and the reasons why the respondent sought to be tested. In this paper, I use logistic regression to determine the odds ratios associated with receiving an HIV test. In addition to several control variables, the independent variables of interest include numerous risk factors known to be associated with HIV infection. The results found indicate that sexual risk factors do not significantly increase the odds of receiving an HIV test. These results speak to the future nature of HIV infection and mortality rates in women.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Most models and theories that aim to understand the spread of infectious diseases tend to simply ignore behavioral changes that occur as the prevalence of an infectious disease increases However, recent empirical findings seem to indicate that sexual risk taking behaviors change as the prevalence of HIV increases. Yet, most research that tries to predict the spread of infectious diseases such as HIV, assumes that actors’ sexual behavior remains constant vis-à-vis changing prevalence rates. This study investigated the effects of two behavioral responses to the HIV epidemic; first, how much HIV transmission changes when actors become more selective about choosing sexual partners and second, the effect of increased condom use on HIV prevalence rates. We constructed a virtual network of 1,000 heterosexual actors and simulated sexual contacts when HIV negative actors were less likely to engage in sexual relations with HIV positive actors and more likely to use condoms and compared the results to a base scenario. We found that assumptions about mixing preferences of HIV negative actors and increased condom use have significant and sizeable impacts on HIV prevalence. Moreover, social isolation may have a larger comparative impact than increased condom use. Lastly, the results of this study suggest that a sociological theory of behavioral responsiveness to sexually transmitted diseases (STDs) needs to be developed.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This presentation investigates behaviors that may be associated with HIV infection among women who use and/or sell crack, powder cocaine and heroin in Central Harlem. Chain referral sampling and other strategies were combined to interview 281 female drug users (N=248) and sellers (N=33) who were tested for HIV infection. Chi-square analysis was used to identify factors associated with HIV infection, and logistic regression analysis was used to ascertain the relative importance of these factors.
More than a fourth (28.5 percent) of all respondents were HIV positive. Thirteen factors were significantly associated with HIV infection in chi-square analyses. In logistic regression analyses, injection drug users were three times more likely to be HIV positive than other women (OR=3.29; 95 percent CI: 1.63—6.66). Latinas (OR=2.84; 95 percent CI: 1.12—7.16), women involved in illegal activities (OR=2.82; 95 percent CI: 1.55—5.14) or women who were separated, divorced or widowed (OR=2.13, 95 percent CI: 1.16—3.90) were more than twice as likely to have HIV. This study adds to previous research that drug injection was the most statistically significant factor associated with HIV infection. However, Latina ethnicity and lifestyle factors that involve income from illegal activities and relationship status were independently associated with HIV infection.