Cost-effectiveness analysis of infant feeding strategies to prevent mother-to-child transmission of HIV in South Africa

Despite increasing availability of perinatal interventions to prevent mother-to-child transmission (MTCT) of HIV in South Africa, MTCT remains high due to breastfeeding. To inform policy decisions in the country, cost-effectiveness of alternative infant-feeding interventions was conducted. Mathematical modelling was used to simulate post-natal transmission and mortality due to infant feeding in a hypothetical cohort of 1 000 HIV-exposed infants. Lifetime costs to the health system were calculated for each strategy. Interventions compared with current practice were: increasing coverage of extended nevirapine prophylaxis (ENP) to infants from 30% (base case) to 60% without changing current feeding practices; actively supporting breastfeeding with ENP to infants for 12 months; and actively supporting exclusive formula (replacement) feeding for 6 months. HIV-free survival at 24 months and disability-adjusted life years (DALYs) averted were estimated for typical rural and certain urban settings. Base-case analysis revealed that expanding coverage of nevirapine prophylaxis with breastfeeding is cost-saving and improves HIV-free survival. Changing feeding practices is beneficial, depending on context. Breastfeeding is dominant (less costly, more effective) in rural settings, whilst formula feeding is a dominant strategy in urban settings. Cost-effectiveness was most sensitive to proportion of women on lifelong antiretroviral therapy (ART) and infant mortality rate (IMR). When >55% of women are on ART, breastfeeding dominates in the urban settings modelled, whilst formula feeding is cost-effective in rural settings when IMR ≤ 45/1000. The study concludes that strategies to support breastfeeding are essential. Strengthening health systems is critical to ensure optimal nevirapine delivery during breastfeeding. A case can be made for formula feeding or breastfeeding in HIV-infected women in specific contexts.     

Effects of a single large dose of vitamin A, given during the postpartum period to HIV-positive women and their infants, on child HIV infection, HIV-free survival, and mortality

BACKGROUND: Low maternal serum retinol level is a risk factor for mother-to-child transmission (MTCT) of human immunodeficiency virus (HIV). Multiple-large-dose vitamin A supplementation of HIV-positive children reduces mortality. The World Health Organization recommends single-large-dose vitamin A supplementation for postpartum women in areas of prevalent vitamin A deficiency; neonatal dosing is under consideration. We investigated the effect that single-large-dose maternal/neonatal vitamin A supplementation has on MTCT, HIV-free survival, and mortality in HIV-exposed infants. METHODS: A total of 14,110 mother-infant pairs were enrolled < or =96 h after delivery, and both mother and infant, mother only, infant only, or neither received vitamin A supplementation in a randomized, placebo-controlled trial with a 2 x 2 factorial design. All but 4 mothers initiated breast-feeding. A total of 4495 infants born to HIV-positive women were included in the present analysis. RESULTS: Neither maternal nor neonatal vitamin A supplementation significantly affected postnatal MTCT or overall mortality between baseline and 24 months. However, the timing of infant HIV infection modified the effect that supplementation had on mortality. Vitamin A supplementation had no effect in infants who were polymerase chain reaction (PCR) positive [corrected] for HIV at baseline. In infants who were PCR negative at baseline and PCR positive at 6 weeks, neonatal supplementation reduced mortality by 28% (P=.01), but maternal supplementation had no effect. In infants who were PCR negative at 6 weeks, all 3 vitamin A regimens were associated with ~2-fold higher mortality (P< or =.05). CONCLUSIONS: Targeted vitamin A supplementation of HIV-positive children prolongs their survival. However, postpartum maternal and neonatal vitamin A supplementation may hasten progression to death in breast-fed children who are PCR negative at 6 weeks. These findings raise concern about universal maternal or neonatal vitamin A supplementation in HIV-endemic areas.     

Breastfeeding to prevent HIV transmission in infants: balancing pros and cons

PURPOSE OF REVIEW: In industrialized countries avoidance of breastfeeding is among the cornerstones of prevention of mother-to-child transmission of HIV. Breastfeeding carries risk for HIV transmission but improves survival, whereas formula feeding carries zero risk for transmission but increased risk for mortality. We assesses breastfeeding is a rational and viable option for HIV-infected women in poor environments. RECENT FINDINGS: A number of recent studies, mostly from Africa, have provided new data that enable health workers to offer HIV-positive women clear advice on infant feeding appropriate to their individual circumstances. The studies are grouped according to whether the evidence favours one or other feeding type and are considered under the following headings: equivalence of formula feeding and breastfeeding; breastfeeding, HIV disease progression and mortality in mothers; breast superior to formula; breastfeeding for HIV-infected babies; and reducing risk for transmission while breastfeeding. SUMMARY: The weight of current evidence favours exclusive breastfeeding for 6 months to prevent mother-to-child HIV transmission for most HIV-infected mothers globally, most of whom live in poor communities; exclusive breastfeeding may also benefit HIV-infected babies. Formula feeding appears to be equivalent to breastfeeding in terms of survival and transmission risk during the first 2 years of life in some settings.     

Mother-to-Child Transmission of HIV and HIV-Free Survival in Swaziland: A Community-Based Household Survey

In Swaziland, no data are available on the rates of HIV infection and HIV-free survival among children at the end of the breastfeeding period. We performed a national crosssectional community survey of children born 18–24 months prior to the study, in randomly selected constituencies in all 4 administrative regions of Swaziland, from April to June 2015. Mother-to-child transmission (MTCT) of HIV and HIV-free survival rates were calculated for all HIV-exposed children. The overall HIV-free survival rate at 18–24 months was 95.9% (95% CI 94.1–97.2). The estimated proportion of HIV infected children among known HIV-exposed children was 3.6% (95% CI 2.4–5.2). Older maternal age, delivering at a health facility, and receiving antenatal antiretroviral drugs were independently associated with reduced risk for child infection or death. The Swaziland program for prevention of MTCT achieved high HIV-free survival (95.9%) and low MTCT (3.6%) rates at 18–24 months of age when Option A (infant prophylaxis) of the WHO 2010 guidelines was implemented.     

“I did not feel like a mother”: The success and remaining challenges to exclusive formula feeding among HIV-positive women in Brazil

Exclusive and safe formula feeding can eliminate the risk of vertical HIV transmission due to breastfeeding. Therefore, many countries advise all HIV-positive women to avoid breastfeeding their infants. However, little research explores the experiences of women attempting to exclusively formula feed in countries with free and universal access to highly active antiretroviral therapy (HAART). This article examines the success of Brazil in supporting HIV-positive women as engage in exclusive formula feeding (EFF). We conducted in-depth interviews with 30 HIV-positive women receiving care at the primary facility for HIV/AIDS in Salvador, Brazil about their attitudes and practices related to EFF as well as challenges with adhering to EFF. All interviews were recorded, professionally transcribed and translated, and then analyzed. Our results showed that one woman reported both breastfeeding and formula feeding her infant; all others reported EFF. Postpartum counseling regarding the risk of HIV transmission through breastfeeding was the primary motivation for EFF. Challenges included difficulty reconciling their perceptions that breastfeeding is an important maternal responsibility, trouble accepting that breastfeeding can cause potential to harm their infants, confronting HIV-related stigma associated with EFF, and unexpected financial burdens due to EFF. We conclude that HIV-positive women adhered to national guidelines recommending EFF; this phenomenon has likely contributed to declining rates of vertical transmission in Brazil. Despite this success, many women experienced challenges with EFF. Greater support services may enhance Brazil's success in empowering HIV-positive women and eliminating vertical HIV transmission via breastfeeding.     

Infant feeding practices of HIV-infected and uninfected women in Zimbabwe

We surveyed infant feeding knowledge, attitudes, and practices in Zimbabwe to determine whether knowledge of HIV seropositivity influences infant feeding behavior. Questionnaires were administered to 97 women 1 and 4 weeks postpartum and prospective data on infant feeding practices were collected. Participants were pregnant women who consented to a HIV test. A total of 116 women participated of whom 99 women underwent voluntary HIV counseling and testing (VCT); 17 women agreed to blinded HIV testing but did not opt for VCT. The responses to questionnaires on infant feeding practices of HIV-positive and HIV-negative women who knew and did not know their HIV status at day 1 and week 4 postpartum were compared. We found that HIV-positive women who did not learn their status breastfed their infants less, introduced supplementary foods sooner, and planned to wean their babies earlier compared to other women (p = 0.005, p = NS, p= 0.02). HIV-positive women (30/97) more frequently reported a prior history of infant death and AIDS-related symptoms compared to HIV-negative women. We conclude that HIV-positive women who did not know their status made incorrect decisions with respect to infant feeding. These women may have suspected themselves to be HIV-positive and consequently underfed their infants or because these women were more symptomatic may have been less likely to breastfeed; decreased intake may increase the risk for malnutrition. Knowledge of HIV status may influence infant feeding decisions and reveal an urgent need to address infant feeding practices of pregnant women in Zimbabwe.     

A feeding education program to prevent mother-to-child transmission of HIV in Haiti

In Haiti, as in most of the developing world, vertical transmission of HIV from infected mother to infant through postpartum breastfeeding remains a significant mode of transmission. As part of their prevention of mother-to-child transmission program, the Groupe Haitien d'Etude du Sarcome de Kaposi et des Infections Opportunistes (GHESKIO) Centers developed a feeding education program in which over 83% of the HIV-positive pregnant women who were eligible to participate, enrolled. Bivariate and adjusted multivariate logistic regression analyses were used to compare feeding choices of the 290 women who participated in the feeding education program to 58 who did not. Of those who participated, 91.7% chose to use replacement formulas for their newborns, while 75.9% of those who did not participate chose replacement feeding. After adjustment for socio-demographic variables, analyses revealed that the no education group was less likely to adopt replacement feeding and more likely to use mixed feeding (OR=0.31, p=0.004; and OR=2.74, p=0.05, respectively). This suggests that a targeted and culturally appropriate education program can be effective in encouraging replacement feeding, even in those countries where breastfeeding is the norm.     

Barriers to participation in the prevention of mother-to-child HIV transmission program in Gaborone, Botswana a qualitative approach

The existing belief that if money and free antiretroviral drugs were offered to developing countries, people would flock to use the services does seem to hold for the Prevention of Mother-to-Child Transmission of HIV Program (PMTCT) in Botswana. The government offers free counseling, testing, free antiretroviral drugs and free infant formula for babies born to infected mothers and yet some women are reluctant to enroll in this program to protect their infants. A qualitative study on factors that motivated and hindered pregnant women from participating in the PMTCT program was conducted in Gaborone, Botswana. Forty subjects (ten HIV-positive pregnant women who accepted the program, eleven who rejected the program, nine PMTCT health workers and ten key informants) participated in in-depth interviews. Thematic content analysis yielded the following barriers to participation in the program: fear of knowing one's own HIV status, infant feeding distribution stigma, lack of male partners' support and negative attitudes of health workers. Reinforcing factors were: availability of free antiretroviral drugs and free infant formula. In conclusion, stigma and discrimination are still the main impediments to women's participation in the PMTCT program. Mitigating these barriers could reduce AIDS-related infant mortality in this country.     

HIV and infant feeding in resource-rich settings: considering the clinical significance of a complicated dilemma

With advances in the care of HIV-positive pregnant women, the likelihood of perinatal transmission is now less than 1%. In resource-rich settings women are instructed to abstain from breastfeeding, as studies have shown that breastfeeding increases the likelihood of infant acquisition of HIV. As practitioners caring for HIV-positive parents, we are now facing growing tension about the complex issues that inform decisions about infant feeding. In the face of changing guidelines and global immigration patterns, simply telling women that breastfeeding is contraindicated may no longer be good enough. We must fully open the lines of communication regarding this important and evolving issue. This commentary will review the clinical, social and cultural considerations that impact decisions regarding infant feeding in the context of HIV.     

Low adherence to recommended infant feeding strategies among HIV-infected women: results from the pilot phase of a randomized trial to prevent mother-to-child transmission in Botswana.

Little is known about the ability of women to adhere to recommended feeding strategies to prevent mother-to-child HIV transmission (MTCT) from breast milk. We conducted a pilot study in rural Botswana to prevent MTCT from breast milk. Women were randomized to formula feed their infants or to exclusively breastfeed while providing prophylactic zidovudine. Women who chose to formula feed independently were also followed. Among those with > or = 3 postpartum visits, none of 31 women assigned to breastfeed did so exclusively for 5 months. Seven (22%) of 32 women in the formula arm definitely or probably breastfed by self-report or as witnessed in maternity, and evidence of breast milk on physical examination was present in 50% of women in > or = 2 visits beyond 1 month. Three (18%) of 17 women choosing formula definitely or probably breastfed, and breast milk was present on exam in 53%. We conclude that adherence to 5 months of exclusive breastfeeding did not occur, and that adherence to exclusive formula feeding was sub-optimal and potentially over reported. Breast examination may be a useful adjunct to self-report, but needs to be validated and standardized. Low adherence to infant feeding strategies that differ from local norms will reduce their effectiveness in preventing MTCT.     

The infant feeding choices and experiences of women living with HIV in Kinshasa, Democratic Republic of Congo

HIV transmission through breastfeeding is a significant public health challenge. While breastfeeding provides important nutrition, and results in reduced morbidity and mortality, there is a risk of HIV transmission through breastfeeding. International prevention of mother-to-child transmission (PMTCT) guidelines recommend exclusive breastfeeding for six months among HIV-infected women on antiretroviral therapy. Promoting exclusive feeding has proved difficult in settings where mixed feeding is a cultural norm. Understanding the factors that influence HIV infected women's infant feeding choices and practices is critical to promoting adherence to PMTCT guidelines. We conducted in-depth interviews with 40 HIV+ pregnant and post-partum women in Kinshasa, Democratic Republic of Congo to understand their infant feeding experiences. Interviews were conducted in Lingala, and transcribed and translated into French for analysis. Deductive and inductive codes were applied, and matrices were created to facilitate cross-case analysis. Women had limited understanding of the specific mechanisms through which their infant feeding practices influenced HIV transmission risk. Clinical staff was the primary source of women's knowledge of HIV mother-to-child-transmission. Among the 24 post-partum women in the sample, seven women adhered to exclusive breastfeeding and two women to exclusive formula feeding for at least six months. Women's beliefs and awareness about HIV transmission through breastfeeding, as well as the information and support from clinical staff and other members of their support networks positively influenced their exclusive feeding. Common barriers to exclusive feeding included financial constraints, breast health problems, misinformation about HIV transmission, local norms, and prior feeding experiences. Health care workers play a key role in providing correct information on PMTCT and supporting women's infant feeding choices to adhere to guidelines of exclusive infant feeding. Optimizing provider-patient communication and creating a supportive environment surrounding infant feeding is critical.     

Challenges faced by health-care providers offering infant-feeding counseling to HIV-positive women in sub-Saharan Africa: a review of current research

Exclusive breastfeeding (EBF) has been identified as the optimal nutrition and critical behavior in attaining human immunodeficiency virus (HIV)-free infant survival in resource-limited settings. Health-care providers (HCPs) in clinic- and community-settings throughout sub-Saharan Africa (sSA) provide infant-feeding counseling. However, rates of EBF at 6 months of age are suboptimal. HCPs are uniquely positioned to educate HIV-positive mothers and provide support by addressing known barriers to EBF. However, limited evidence exists on the experiences faced by HCPs in providing counseling on infant feeding to HIV-positive women. Our objective is to describe experiences faced by HCPs when delivering infant-feeding counseling in the context of HIV in program settings in sSA. We searched a range of electronic databases, including PubMed, CINAHL, and PsycINFO from January 1990 to February 2013, in addition to hand-searching, cross-reference searching, and personal communications. The search was limited to publications in English. Empirical studies of HCP experiences providing infant-feeding counseling in the prevention of mother-to-child transmission (PMTCT) of HIV programs in sSA were selected. We identified 10 peer-reviewed articles reporting HCP challenges in infant-feeding counseling that met inclusion criteria. Articles included qualitative, cross-sectional and mixed-method studies, and cumulatively reported 31 challenges faced by HCPs. Among the challenges identified, the most commonly reported were personal beliefs held by the HCPs toward infant feeding in the context of HIV, contradictory messages, staff workload, directive counseling styles, and a lack of practical strategies to offer mothers, often leading to improvised counseling approaches. Counseling strategies need to be developed that are relevant, meaningful, and responsive to the needs of both HCPs and mothers.     

Integrating prevention of mother-to-child HIV transmission into antenatal care: learning from the experiences of women in South Africa

In 1999, for the first time in South Africa, a Mother-to-Child HIV Transmission (MTCT) prevention programme was implemented at the routine primary care level and not as part of a research protocol. A total of 264 women attending prenatal care in these clinics were interviewed in Xhosa using a standardized questionnaire. All had been offered HIV testing, and 95% had accepted. Women who had not been tested were four times more likely to believe that in the community families reject HIV-positive women (p<0.005). Of women who tested, 19% were HIV positive and 83% had told their partner that they had taken the test. HIV-positive women who had not disclosed testing to their partners were three times more likely to believe that, in the community, partners are violent towards HIV-positive women (p<0.005); 86% stated that they would have taken AZT if found to be HIV positive. Only 11% considered that the use of formula feeding indicated that a woman was HIV positive. In conclusion, routine prenatal HIV testing and interventions to reduce perinatal HIV transmission are acceptable to the majority of women in a South African urban township, despite an awareness of discrimination in the community towards HIV-positive women.     

Breast-feeding guidelines: an implementation puzzle. Experts say there is no easy solution

The WHO is attempting to provide guidance in the thorny dilemma faced by many HIV-positive women: should they breast-feed and risk transmission to their infant, or should they use substitute nutrition which may place their infant at greater risk of dying within the first year?     

Infant feeding practices and attitudes among women with HIV infection in northern Thailand

Knowledge and attitudes towards infant feeding among women in northern Thailand were examined. Face-to-face interviews using structured questionnaires were undertaken in three districts of Chiang Rai province. Subjects included postnatal women with HIV infection (group 1, n=80), antenatal women with HIV infection (group 2, n=36) and antenatal women with unknown HIV status (group 3, n=86). Advantages of breastfeeding and formula feeding according to several characteristics (convenience, cleanliness, cheapness and safety) were rated using a four-point (0-3) scale. Overall, breastfeeding was rated much higher (11.4/12) than formula feeding (6.1/12)(p < 0.0005). Formula feeding rating was highest among postnatal women with HIV infection (6.8/12); however, it was lower than the rating for breastfeeding (11.3/12). The vast majority of women with HIV infection were either formula feeding (group 1, 94%) or intended to formula feed (group 2, 72%) their infants. In contrast, the vast majority of antenatal women of unknown HIV status planned to breastfeed (group 3, 83%). All women, regardless of HIV status, consider breastfeeding to be more advantageous than formula feeding. However, once women with HIV infection are informed of the risk of HIV transmission through breastfeeding, they are able to make their own decision to follow the Thai Ministry of Public Health's recommendation to formula feed.     

河南省2005-2007年孕产妇HIV检测情况分析

目的了解河南省孕产妇艾滋病病毒（HIV）感染状况,为开展预防艾滋病母婴传播工作提供决策依据。方法对2005年1月-2007年5月,河南省艾滋病疫情相对集中的31个县（区）孕产妇的HIV检测结果进行分析。结果共对31个艾滋病重点县720605名孕妇和产妇检测HIV抗体,检测率93.11%;检测出HIV抗体阳性541例,阳性率0.08%。其中364例为阳性孕妇,223例自愿终止妊娠（占61.27%）,5例失访（1.37%）,136例选择继续妊娠待分娩（占37.36%）。3年间选择继续妊娠的HIV阳性孕妇,分别为31.51%（46/146）、36.31%（57/157）、54.10%（33/61）,呈逐年上升趋势。284例分娩阳性产妇中,250例（88%）采取了母婴阻断预防措施。结论加强孕产妇HIV检测和阳性育龄妇女的孕检工作,早期发现、早期管理HIV阳性孕妇,可以进一步提高HIV阳性产妇的药物阻断率。     

Promoting Condom Use With Main Partners: A Behavioral Intervention Trial for Women

The Stages of Change (SOC) model was used to implement and evaluate a condom promotion intervention for HIV-positive and at-risk women who were recruited from clinic and community settings in Baltimore and Philadelphia. Participants were assigned to receive standard reproductive health services or enhanced services (standard plus SOC peer advocate intervention). Women who had a main partner at baseline and at a 6-month follow-up were included in the analysis (70 HIV positive, 471 at risk). Compared with the standard group, HIV-positive women in the enhanced group were significantly more likely to have progressed in the SOC model or to have maintained consistent condom use, and less likely to have relapsed or stayed in the precontemplation stage. Among the at-risk women, exposure to the intervention was associated with being at a higher SOC and being less likely to relapse relative to the standard group. The SOC model has promise for use in social and public health service settings that serve women at risk or living with HIV.     

Method of feeding and transmission of HIV-1 from mothers to children by 15 months of age: prospective cohort study from Durban, South Africa

OBJECTIVE: To determine the risk of HIV transmission by infant feeding modality. DESIGN AND SETTING: A prospective study in two hospitals in Durban, South Africa. PARTICIPANTS: A total of 551 HIV-infected pregnant women enrolled in a randomized trial of vitamin A. INTERVENTIONS: Women self-selected to breastfeed or formula feed after being counselled. Breastfeeders were encouraged to practice exclusive breastfeeding for 3-6 months. MAIN OUTCOME MEASURES: Cumulative probabilities of detecting HIV over time were estimated using Kaplan-Meier methods and were compared in three groups: 157 formula-fed (never breastfed); 118 exclusively breastfed for 3 months or more; and 276 mixed breastfed. RESULTS: The three feeding groups did not differ in any risk factors for transmission, and the probability of detecting HIV at birth was similar. Cumulative probabilities of HIV detection remained similar among never and exclusive breastfeeders up to 6 months: 0.194 (95% CI 0.136-0.260) and 0.194 (95% CI 0.125-0.274), respectively, whereas the probabilities among mixed breastfeeders soon surpassed both groups reaching 0.261 (95% CI 0.205-0.319) by 6 months. By 15 months, the cumulative probability of HIV infection remained lower among those who exclusively breastfed for 3 months or more than among other breastfeeders (0.247 versus 0.359). CONCLUSION: Infants exclusively breastfed for 3 months or more had no excess risk of HIV infection over 6 months than those never breastfed. These findings, if confirmed elsewhere, can influence public health policies on feeding choices available to HIV-infected mothers in developing countries.     

不同用药方法阻断艾滋病母婴传播73例临床观察

目的 观察不同用药方法 阻断艾滋病母婴传播效果.方法 2005~2007年应用维乐命方案(方案1)阻断HIV母婴传播26例,2008~2010年联合应用三联抗病毒药物方案(方案2)阻断HIV母婴传播47例.所出生的婴儿均采用人工喂养,随访观察至18月龄.结果 两种方案出生的婴儿外观均未见畸形,方案1出现2例婴儿HIV阳性(母儿均服药及仅婴儿服药各1例);方案2出现1例婴儿HIV阳性(仅婴儿服药),母儿均服药均为阴性.结论 对HIV阳性孕妇采取综合干预措施(抗病毒药物治疗、婴儿人工喂养)能有效阻断HIV母婴传播.     

HIV, malaria, and infant anemia as risk factors for postneonatal infant mortality among HIV-seropositive women in Kisumu, Kenya

BACKGROUND: HIV and malaria in sub-Saharan Africa are associated with poor pregnancy outcome and infant survival. We studied the association of placental malaria, infant malaria and anemia, and infant HIV status with postneonatal infant mortality (PNIM) among infants of HIV-seropositive women. METHODS: During 1996-2001, infants born to 570 HIV-seropositive mothers in Kisumu, Kenya were monitored monthly for malaria (parasitemia or clinical malaria) and anemia (hemoglobin level <8 g/dL) and vital status. RESULTS: Thirty-nine deaths occurred among 112 HIV-positive infants (420/1000 live births [LBs] [95% confidence interval {CI}, 318-522 LBs]), and 36 occurred among 458 HIV-negative infants (99/1000 LBs [95% CI, 68-130 LBs]) (P<.001). In multivariate Cox regression analysis among HIV-negative infants, PNIM was associated with infant anemia (adjusted hazard ratio [AHR], 5.03 [95% CI, 1.97-12.81]) but not with placental malaria (AHR, 1.22 [95% CI, 0.50-2.95]) or infant malaria (AHR, 0.35 [95% CI, 0.10-1.21]). Among HIV-positive infants, neither placental malaria (AHR, 0.34 [95% CI, 0.10-1.10]) nor infant malaria (AHR, 0.31 [95% CI, 0.07-1.33]) or anemia (AHR, 1.07 [95% CI, 0.32-3.61]) was significantly associated with PNIM. CONCLUSION: In this study population, placental malaria and infant parasitemia were not risk factors for PNIM among infants of HIV-seropositive women. The prevention of infant anemia may decrease PNIM among HIV-negative infants of HIV-seropositive women.