February 4, 2005
CDC and NCDOH reviewed public health surveillance data for 1998-2004 for newly reported HIV infections and HIV contact-tracing records of NC Disease Intervention Specialists (NCDIS). A case-control study was conducted to identify epidemiologic and behavioral differences between HIV-positive women (case-patients) and HIV-negative women (controls). Case-patients and controls were heterosexually active, non-injection-drug using black women aged 18-40 years who resided in regions with highest HIV morbidity (i.e., Raleigh, Durham, and Charlotte). Case-patients received an HIV diagnosis during January 2003 - August 2004 and were recruited through NCDIS and medical clinics. Controls were recruited at HIV-testing sites during August-October 2004. Attempts were made to interview the HIV-positive male sex partners of case-patients to assess male partner risk factors. Face-to-face interviews were conducted with all participants to obtain epidemiologic and behavioral information, including sexual behaviors reported for the 12-month period preceding either the date of diagnosis for the case-patients and their partners or the date of interview for the controls. Participants were also asked to offer explanations for HIV risk behaviors among black women and to provide suggestions for strategies to reduce HIV transmission among black women in North Carolina. Univariate and multivariate analyses* were conducted by using statistical analysis software. Unless otherwise noted, all differences indicated in this report are statistically significant at p<0.05.
During January 2003 - August 2004, a total of 208 new HIV infections were reported among black women aged 18-40 years in Raleigh, Durham, and Charlotte. Of these, 97 (47%) patients were excluded, including 75 whose HIV was diagnosed before 2003 or in another state, 13 who could not be located, and nine who were deceased or too ill to be interviewed. Of the remaining 111 patients, 31 (28%) agreed to participate in the interview; 58 (52%) could not be located or did not respond to inquiries, and 22 (20%) refused participation. A total of 101 controls agreed to participate in the interview. Controls were recruited during HIV testing at health departments (87 controls), at an apartment complex (nine), at a pharmacy (three), and at a church (two). Of the 31 patients, 15 (48%) could identify an HIV-positive male who they suspected was the source of their HIV infection. Of these men, six (40%) agreed to be interviewed.
Patients and controls were demographically and behaviorally similar (Table). Similarities included their median age at first sexual intercourse, lifetime histories of any sexually transmitted disease (STD), reports of unprotected vaginal intercourse, and previous HIV testing. Although the majority of participants had either previously had an STD, been pregnant, or been tested for HIV, most felt they were unlikely or very unlikely to contract HIV. Seven (23%) patients learned they were HIV-positive during prenatal screening.
In the univariate analysis, several differences were identified among patients and controls (Table). Patients were significantly more likely than controls to be unemployed; have 20 or more lifetime sex partners; use crack/cocaine; and receive money, shelter, or drugs in exchange for sex. In the multivariate analysis, three statistically significant differences were identified. Women receiving public assistance (adjusted odds ratio [AOR] = 7.3; 95% confidence interval [CI] = 2.1-26.0) and who had a lifetime history of genital herpes infection (AOR = 10.6; CI = 2.4-47.2) were significantly more likely to be HIV-positive. However, women who discussed sexual and behavioral history with their male partners were less likely to be HIV-positive (AOR = 0.6; CI = 0.4-0.8).
According to NCDIS records, three of the six male partners who agreed to an interview had engaged in sex with another male, but only one admitted to this activity during the interview; none reported injecting drugs. Twenty-two (71%) of the HIV-positive women believed they were infected by a steady partner. Although only one third of the HIV-positive women characterized the relationship with their steady partner as mutually monogamous, the most common reason reported for not using condoms was that they trusted their partners.
The most common reasons reported by black women for engaging in behaviors that place them at risk for HIV infection were 1) financial dependence on male partners, 2) feeling invincible, 3) low self-esteem coupled with a need to feel loved by a male figure, and 4) alcohol and drug use. In addition, participant's proposed strategies for reducing HIV transmission among black women in North Carolina included 1) introducing HIV and STD educational activities in elementary and middle schools, 2) increasing condom availability and usage, and 3) integrating targeted HIV-education and -prevention messages into church and community activities, as well as into media and popular culture.
Reported by: P Leone, M.D., A Adimora, M.D., Univ of North Carolina, Chapel Hill; E Foust, M.P.H., D Williams, Ph.D., M Buie, M.A., J Peebles, North Carolina Dept of Health. L Fitzpatrick, M.D., E McLellan-Lemal, M.S., W Chege, M.D., JT Brooks, M.D., G Marks, Ph.D., S Knox, M.P.H., M Williams, Ph.D., A Greenberg, M.D., Div of HIV/AIDS Prevention, National Center for HIV, STD, and TB Prevention; F Forna, M.D., EIS Officer, CDC.
Results of the multivariate analysis provide insights into developing prevention strategies for black women. Although some HIV-positive women might have been receiving public assistance related to their HIV infection, the findings of higher rates of public assistance and exchanging drugs, money, or gifts for sex among HIV-positive women are consistent with previous studies, which suggested that economic disparities might contribute to the HIV epidemic among black women.3,4 A history of genital herpes was also strongly associated with HIV infection. This finding supports the need for intervention programs offering comprehensive diagnosis, treatment, and prevention services for HIV and other STDs.5 In addition, HIV-positive women were less likely to discuss HIV risk behaviors with their male sex partners. Encouraging women to discuss with their partners HIV status and other STD and drug-use history might provide them with information that leads to HIV risk reduction.
Because some study participants had been sexually active at a young age, targeted HIV-prevention messages might be more effective if introduced at younger ages and widely disseminated through various modalities, including families and those channels suggested by study participants. Furthermore, because many HIV-positive women were unaware of their HIV status until after presentation for prenatal care, integration of routine HIV testing and prevention messages in medical settings for sexually active persons might be beneficial.
The findings in this report are subject to at least five limitations. First, the low participation rates among patients might have introduced selection bias. Data comparing participants with nonparticipants were not available. Second, the results might not be applicable to all black women at risk for acquiring HIV infection, particularly those who are of higher socioeconomic status. Third, assessment of many of the complex sociocultural factors that likely influence HIV risk in this population was not possible. Fourth, causality could not be demonstrated in the association between HIV and a history of herpes, and the relationship between receipt of public assistance and HIV serostatus could not be clarified. Finally, because only a few HIV-positive male partners were interviewed in this investigation, describing the role of male partner risk in HIV transmission among black women in North Carolina was not possible.
Findings from this investigation demonstrate the need for multiple approaches to reducing HIV infection among black women. CDC, in collaboration with state and local health departments and community-based organizations (CBOs), is disseminating effective HIV-prevention interventions that target sexually active black women.6,7 In addition, CDC's Advancing HIV Prevention strategy has introduced programs (e.g., HIV testing and sexual network demonstration projects) to improve HIV testing in at-risk populations.8 Whereas CBOs are funded to implement these programs throughout the United States, ensuring that such programs are accessible to black women living in disadvantaged areas of the urban and rural South is vital. In addition, more resources and prevention strategies are needed to help address underlying causes of HIV transmission in black women, such as poverty and partner risk behavior. Halting the spread of HIV among black women will require HIV-prevention strategies and programs that encourage delays in sexual activity, consistent condom use, mutually monogamous relationships, and improved partner communication. Furthermore, efforts are needed to introduce age-appropriate sex education before beginning of sexual activity, improve the availability of HIV and STD testing and treatment, and focus attention to the economic constraints that create challenges for disadvantaged black women to prioritize health issues such as HIV.
* All variables with p<0.1 in the univariate analysis were included in the multivariate model.