Among HIV-positive women participating in an HIV prevention program, the odds of improving consistency of condom use with a main partner are more than twice as high for those who have access to peer advocates as for those who do not.1 Women's odds of being confident in their ability to use a condom every time they have sex with their main partner are doubled if they have access to peer advocates. However, the same intervention has little effect on condom use among women at risk of HIV infection. These findings are based on an evaluation involving 1,611 women in Baltimore and Philadelphia.
Between March 1993 and September 1995, researchers recruited 322 HIV-positive, nonpregnant women aged 18-44 in Baltimore and 1,289 nonpregnant at-risk women aged 15-44 in Philadelphia to participate in an HIV prevention program. In Baltimore, women were recruited from a hospital-based oupatient HIV or pediatric HIV clinic, from a community-based primary HIV care facility or through informal referrals from outreach workers. Women's HIV status was confirmed through medical records, if available, or through an HIV test. In Philadelphia, women were recruited from drug treatment facilities, homeless shelters and public housing developments. These facilities were chosen because their clients tend to be at risk of HIV infection (for example, because they or their partners are injection-drug users, or they exchange sex for money or drugs).
The six-month interventions were the same in both cities: Women participated in either an HIV prevention program that provided access to comprehensive reproductive health care services (standard services) or a program that provided the same services accompanied by access to peer advocates (enhanced services). Peer advocates focused on three behaviors: condom use with main male partners, condom use with other male partners and contraceptive use.
Tailoring services to each client's needs, peer advocates provided an intervention designed to help women move along a continuum of behavior change, from "precontemplation" (signifying that the woman was not yet considering a new behavior) to "maintenance" (meaning that she had mastered and sustained a new behavior). Women could meet one-on-one with advocates as many times as they wanted and could attend weekly group sessions.
Women in Baltimore were randomly assigned to an intervention group; 158 women received standard services and 164 received enhanced services. In Philadelphia, assignment to a group was based on recruitment site, because it was not possible to randomly assign women within facilities; 566 women received standard services, and 723 women received enhanced services.
Women completed a baseline interview and follow-up interviews at six, 12 and 18 months. Interviewers asked about women's demographic and risk characteristics; condom and other contraceptive use with main and other partners; confidence in their ability to use a method (or self-efficacy); and perceptions of advantages and disadvantages of use.
On the basis of their responses to the questions about condom and other contraceptive use, women were categorized into one stage along the continuum of behavior change. If women moved forward or backward along the continuum from one interview to the next, they were considered to have either progressed or relapsed in a particular behavior. The researchers conducted regression analyses to compare the probabilities of changes in behavior, self-efficacy and perceived advantages in the two treatment groups. The analyses were of 124 HIV-positive women and 843 at-risk women.
The HIV-positive women were similar to those at risk in mean age (32 and 30 years, respectively) and mean length of their relationship with a main partner (five years). Similar proportions in both groups had less than a high school education (49% and 56%), had ever had a sexually transmitted disease (67% and 61%) and had ever exchanged sex for money or drugs (35% and 41%). Larger proportions of women who were HIV-positive than of at-risk women had ever injected drugs (56% vs. 18%), had a main partner who was HIV-positive (43% vs. 2%) and had only one sexual partner (88% vs. 65%). Larger proportions of infected women than of at-risk women were considered to be in the maintenance category of contraceptive use (59% vs. 28%) and of condom use with a main partner (48% vs. 9%) and other partners (33% vs. 20%). By contrast, larger proportions of at-risk women were in the precontemplation or contemplation stages for condom use with main partner (68% vs. 26%) and contraceptive use (41% vs. 29%).
Across all interview periods, among HIV-positive women receiving enhanced services, the odds of progressing in condom use with a main partner were more than twice those of women receiving standard services (odds ratio, 2.3), and the odds of relapsing were less than half those of women receiving standard services (0.4). In addition, for women receiving enhanced services, the odds of reporting confidence in their ability to use condoms with a main partner and of perceiving advantages to such use were twice those of women receiving standard services (2.0 and 1.9, respectively).
However, there were fewer significant differences between the two groups at specific follow-up periods. Only at the six-month interview did women receiving enhanced services have elevated odds of progressing in condom use with a main partner (odds ratio, 2.8) and reduced odds of relapsing (0.3); only at the 12-month interview did they have elevated odds for self-efficacy (7.4). The researchers did not conduct an analysis of women's behavior with other partners because only 19 infected women reported such partners at any of the interviews.
Regarding contraceptive use, overall, women receiving enhanced services were significantly less likely than those receiving standard services to have relapsed (odds ratio, 0.4). There were no significant differences in contraceptive use or perceptions of use between the two groups at six months. At the 12-month follow-up, women receiving enhanced services had elevated odds of perceiving disadvantages of using contraceptives (4.1). At 18 months, they were significantly more likely to have progressed and significantly less likely to have relapsed (4.1 and 0.2, respectively). In addition, they had significantly elevated odds of perceiving advantages of contraceptive use (3.7). The investigators were not able to analyze self-efficacy for contraceptive use because the sample size for this variable was too small.
Among women at risk of HIV infection, receiving enhanced services was associated with small positive effects on behavior and perceptions, and in some cases was associated with the opposite of the desired effects. The only significant overall differences between intervention groups were that women receiving enhanced services had slightly elevated odds of perceiving advantages to using condoms with a main partner (odds ratio, 1.4) and had reduced odds for self-efficacy (0.8).
At the six-month follow-up, women who had received enhanced services were less likely than those getting standard care to report relapsing in their condom use with a main partner (odds ratio, 0.7), and were more likely to report feeling confident in their ability to use a condom every time they had sex with their main partner (1.5). In addition, at the 12-month follow-up they had reduced odds of perceiving disadvantages of contraceptive use. However, at 12 months, women receiving enhanced services were significantly less likely to report self-efficacy in using condoms with a main partner (0.4) or other partner (0.5), or in using contraceptives (0.5). There were no significant differences between the two groups at the 18-month follow-up.
The researchers speculate that there could be several reasons the intervention had different effects for HIV-positive women and women at risk of infection. For example, almost half of the HIV-positive women were in the maintenance stage at the beginning of the study, while more than half of at-risk women were in the precontemplation stage. The enhanced intervention targets these stages with very different counseling activities. For women in the early stages of behavior change, the counseling is very informational and the content is similar to that of standard services. Thus, because there was less of a difference in the intervention and control services most of the at-risk women received compared with what HIV-positive women received, there was less of an effect. Another possibility for the different findings is that women who are HIV-positive may be more highly motivated to use condoms or contraceptives. In addition, according to the researchers, the HIV services they receive may help them maintain a more stable lifestyle than that of women who are in treatment for substance abuse or live in shelters.
The investigators acknowledge several limitations of their research, including the possibility that women in the HIV study, who were seeking medical care, may not be representative of all HIV-positive women. Even so, according to the researchers, "The enhanced intervention's success among HIV-positive women suggests it should be considered among the tools public health professionals use to encourage condom use among HIV-positive women receiving primary HIV care."--B. Brown
1. Fogarty LA et al., Long-term effectiveness of a peer-based intervention to promote condom and contraceptive use among HIV-positive and at-risk women, Public Health Reports, 2001, 116 (Suppl. 1):103-119.