Staying Healthy and Reducing Transmission

A systematic review adhering to the Cochrane Group review guidelines was conducted of observational studies and randomized control trials to evaluate the rates of sexual HIV transmission between heterosexual serodiscordant couples when the HIV-positive partner had full viral suppression on ART.  Of the 3 studies that met all eligibility criteria with confirmed full virologic suppression in the HIV-positive partner, the rate of transmission was 0 per 100 person-years (no seroconversions were reported).  With an additional 3 studies that did not confirm full viral suppression, the rate of transmission was 0.14 per 100 person-years; however, all transmission events occurred shortly following initiation of treatment (before viral suppression could be attained), so excluding these transmission events the rate of transmission was still 0 per 100 person-years.  When converted to lifetime risk of seroconversion, the upper end of the confidence interval (.0001 per 100 person-years) demonstrates a 1 in 204 to 1 in 50 chance of transmission to the infected partner over 20 to 50 additional years of life.  Studies were included that provided data on sexual contact, HIV-positive partner on ART, confirmed undetectable viral load at the time of transmission, and reported seroconversions rates.  Of 20,252 records originally identified, three met all the inclusion criteria.  These three cohort studies were conducted on 991 heterosexual couples in Brazil, Spain and Uganda with 2,064 person-years of follow-up.  Two cohort studies and one randomized control trial were also used in further analysis that met all of the inclusion criteria except the confirmed viral load at the time of transmission.  The cohort studies were conducted in Botswana, Kenya, Rwanda, South Africa, Tanzania, Uganda, and Zambia on 4,307 couples. The randomized control trial was the HPTN 052 trial carried out on 1,763 couples in Botswana, Kenya, Malawi, South Africa, Zimbabwe, Brazil, India, Thailand, and the United States (Cohen et al., 2011a).  There was a total of 8,170 person-years of follow-up in the latter 3 studies.  These studies reported 4 transmission events, all of which occurred early after the initiation of ART.  The study notes that, "not enough data were available to present results on transmission rates through intercourse exclusively without condoms, by type of sexual act (vaginal vs. anal), correcting for the presence of STIs, or by female-to-male or male-to-female contact." (Loutfy et al., 2013: 3). Unprotected sexual intercourse may not result in transmission when viral suppression in the HIV-positive partner has been confirmed.

A Cochrane review of one randomized controlled trial (Cohen et al., 2011a) and nine observational studies conducted in Botswana, Brazil, China, India, Italy, Malawi, Kenya, Rwanda, South Africa, Spain, Thailand, Uganda, the United States of America, Zambia, and Zimbabwe was conducted to determine if ART use was associated with lower risk of seroconversion in HIV-discordant couples.  This review found that in couples where the infected partner (index partner) was being treated with ART, the uninfected partner had at least a 40% lower risk of acquiring HIV than in couples where the index partner was not receiving ART.  The ten studies identified 2,112 episodes of HIV transmission, 1,016 among treated couples and 1,096 among untreated couples.  In the observational studies, the risk for HIV seroconversion incidence in the ART-naïve group was 54 cases per 1,000 people.  The risk for HIV seroconversion incidence in serodiscordant couples where the index partner was on ART was 31 per 1,000 people.  The data was also categorized into subgroups by the index partner’s initial CD4 count.  When the index partner had a CD4 count of more than 350 cells/µl, there was an 88% lower risk of seroconversion among couples where the index patient was on ART.  In this group, there were 247 seroconversions in the ART-naïve group and 30 in the ART group.  This review demonstrates that ART can be used as a preventive measure for HIV transmission among serodiscordant couples, and that this intervention is effective at CD4 counts above 350 cells/µl.

A randomized trial of 1,763 couples in nine countries – Botswana, Kenya, Malawi, South Africa, Zimbabwe, Brazil, India, Thailand and the United States - in which one partner was HIV-1-positive and the other was HIV-negative, with 50% of partners were men living with HIV, found that early initiation of antiretroviral therapy at CD4 counts between 350 and 550 in 886 couples resulted in a 96% relative risk reduction of HIV transmission to the HIV-negative sexual partner as compared to initiation of antiretroviral therapy at CD4 counts at 250 or less in 877 couples. Of the 39 HIV transmissions, 28 were virologically linked to the infected partner. Of the 28 linked transmissions, only one occurred in the early therapy group, for an incidence rate of 0.1 per 100 person years in the early initiation group compared to an incidence rate of 0.9 per 100 person years in the later initiation group. The single HIV transmission event in the early treatment group was ascribed to transmission before HIV suppression was possible (Cohen et al., 2012b). Of the 28 HIV-positive participants who had linked transmission to a partner, 17 (61%) had a CD4 count of more than 350 cells per cubic millimeter at the study visit before the detection of linked HIV transmission. All linked transmissions in the delayed therapy group occurred while the HIV-positive participant was not receiving antiretroviral therapy. There was a relative reduction of 89% in the total number of HIV transmissions resulting from the early initiation of antiretroviral therapy, regardless of viral linkage with the infected partner. Women living with HIV were the source of infection in 18 of 27 (67%) linked transmissions in the delayed therapy group and a man was the source of the single transmission in the early therapy group. HIV-1 uninfected partners were encouraged to return for all visits together for counseling on risk reduction, condom use, and treatment of STIs. Self-reported 100% condom use at baseline was associated with a reduced risk of HIV transmission. Of the couples, 97% were heterosexual and 94% were married. At enrollment, 1,291 of the HIV-positive (73%) reported having had at least one sexual encounter during the previous weekend and 5% had unprotected sex, with a similar profile for the HIV-negative partner. Partners who seroconverted to HIV-positive were released from the study and referred for care. Any woman who was pregnant at enrollment was provided antiretroviral therapy appropriate for use during pregnancy at the start of the second trimester and women in the delayed-therapy group discontinued antiretroviral therapy at delivery or when breastfeeding ended. Twelve new HIV-negative partners who met study criteria for inclusion were enrolled with an HIV-positive partner after the original partner was released from the study.

A study done utilizing behavioral data from the continuing Temprano-ANRS12136 randomized, controlled trial in Côte d'Ivoire compared the sexual behaviors of patients who received early ART as opposed to standard ART. By accounting for sexual behavior and viral loads, the study authors did calculate that the protective effect of early ART was 89% for sexually active people, which “was mainly attributable to difference in viral loads between patients receiving early versus standard ART,” (Jean et al., 2014: 435).  The study found that the early ART patients reported slightly less risky sex in the past month; however, the difference was not statistically significant.  So although the sexual behaviors between the standard and early ART groups was similar, ART significantly reduced the chance of transmission at each sexual encounter.  HIV-positive adults were enrolled in this trial from 2008 to 2012 if they had a CD4 count of less than 800 cells/µl but did not satisfy any criteria for initiating ART.  Participants were randomly assigned to initiate ART immediately (early ART group), or to delay treatment until they met one WHO-recommended criterion for initiating ART (standard ART group).  WHO-recommended initiation criteria was below CD4 count below 200 cells/µl before 2010, when it changed to below 350 cells/µl.  A questionnaire was completed by 957 participants at their 12-month visit.  This included data on the last episode of sexual intercourse (past month or past year), including type of partnership (cohabiting or not), and partner’s HIV status (unknown, negative, or positive).  Risky sex was defined as an instance of unprotected intercourse with a HIV-negative or HIV-unknown partner.  The partner’s exposure was defined as risky if the sexual encounter was unprotected and the viral load of the HIV-positive partner’s CD4 count was more than 300 cells/µl.  At baseline, 80.4% of participants were women, the median age was 35 years old, 46.2% were cohabiting, and the median CD4 count was 478 cells/µl.  By the 12-month visit, 15% of the standard ART group had initiated treatment.  Risky sex was reported by 10.0% of the early ART group and 12.8% of the standard ART group. The proportion of participants exposing their partners to HIV infection (measured by viral load) was 2.4% in the early ART group and 10.7% in the standard ART group.  Of the participants, 46.0% reported sexual activity within the past month, and 41.5% of that activity was between noncohabiting partners.  The early ART group reported that their last sexual partner was HIV-negative in 22.8% of cases and that their status was unknown in 47.7% of cases.  The standard ART group reported that their last sexual partner was HIV-negative in 26.6% of cases and that their status was unknown in 43.9% of cases.  The estimated transmission rate at the last episode of sexual intercourse was 4.0 cases per 10,000 sexually active persons in the standard ART group and 0.5 cases per 10,000 sexually active persons in the early ART group.  The corresponding estimated protective effect of early ART against HIV transmission was 89%.  However, the study authors note that, “the social acceptability and equity of prioritizing access to early ART to this population is questionable,” (Delva et al., 2012 cited in Jean et al., 2014: 438).  With this data, the study authors estimate that early ART could prevent 161 infections per 10,000 patients in the first year of treatment.

A prospective longitudinal study was conducted in Yunnan province in China from 2009 to 2011 to assess the efficiency of China’s current HIV treatment program to prevent new HIV infections among discordant couples in rural China.  In total, 813 couples were followed for an average of 1.4 person-years. The 288 couples that were lost to follow-up had a greater proportion of HIV-positive partners that were not receiving antiretroviral therapy.  Routine ART was prescribed according to China’s national guidelines (CD4 count less than 350 cells/µl).  During the study, the couples were classified as ART-experienced or ART-naïve.  The HIV-positive partners who started ART before the beginning of the study and those who had started ART before the midpoint of the follow-up period were classified as ART-experienced.  The remaining couples were classified as ART-naïve.  Each partner was interviewed at baseline and at the follow-up visit.  Samples were taken of each partner’s plasma at each visit to test for HIV, syphilis, herpes, CD4 count, and viral load testing.  The HIV-negative partners were 79.5% women, 70.2% were aged 39 years and younger, 68.9% were illiterate or educated at primary school only, and 96.8% had never used drugs.  A reported 48.1% of HIV-positive partners had a history of drug use, and 47.8% of HIV-positive partners were on ART at the baseline interview.  HIV-positive partners on ART were more likely to have a viral load of less than 400 copies/ml (as compared to HIV-positive partners not on ART).  A total of 17 seroconversions were recorded during 1,127 person-years, with an overall incidence of 1.5 seroconversions per 100 person-years.  HIV incidence was higher (9.7 per 100 person-years) among participants who reported inconsistent condom use.  HIV incidence was also higher for partners with a viral load greater than 400 copies/ml (2.3 per 100 person-years), or for partners who were not receiving ART (2.4 per 100 person years).  This data, "suggests that HIV-negative spouses of HIV-positive patients are a…high-risk population," (He et al., 2013: 9).  Five of the 17 seroconversions occurred in couples in which the HIV-positive partner was on ART prior to baseline.  All of these couples reported inconsistent condom use and four HIV-positive partners had a viral load greater than 10,000 copies/ml at baseline, "suggesting that virological failure may have resulted in these transmission events," (He et al., 2013: 9).  This data shows that when the HIV-positive partner is on ART, the couple has one-third the risk of seroconversion as compared to couples where the HIV-positive partner is not on ART. The population on ART had a seroconversion rate of 0.8 per 100 person-years, a 66% reduction from the 2.4 per 100 person-years in the ART-naïve population.

A retrospective cohort study in China from 2003 to 2011 was conducted to measure the effects of ART on transmission of HIV in serodiscordant couples.  The ART-naïve cohort had a rate of transmission of 2.6 per 100 person-years and the ART-treated cohort had a rate of transmission of 1.3 per 100 person-years, demonstrating a 26% relative risk reduction in HIV transmission.  ART was initiated in all HIV-positive individuals who met Chinese treatment criteria (when CD4 count dropped below 200 cells/µl until 2008, when guidelines changed to CD4 count dropping below 350 cells/µl).  Repeat tests for HIV for HIV-negative partners and CD4 counts for HIV-positive partners were recommended every 6 months.  Couples were taken from the databases and categorized as the treatment-naïve cohort or the treatment cohort.  From the databases, 24,057 couples were included in the treatment cohort and 14,805 couples were included in the treatment-naïve cohort.  A total of 1,631 seroconversions were recorded from 2003 to 2011.  Treated patients generally had the same demographic characteristics as treatment-naïve patients; however, treated patients were on average older, were predominantly infected through blood or plasma transfusion, and had a longer follow-up time.  Treatment-naïve patients were predominantly infected through sexual contact.  Treatment was the most protective in the first year of follow-up, but not in subsequent years.  Treatment was significantly protective when the HIV-positive partner had been infected through transfusion or heterosexual intercourse, but was not significantly protective when they had been infected through male homosexual sexual contact or by injecting drugs. The study also found that women were more likely than men to transmit HIV to their partner.

A review of publications from 1996 to 2009 with 11 cohorts reporting on 5,021 heterosexual couples and 461 HIV transmission events found that studies of heterosexual discordant couples observed no transmission in patients treated with ART and with viral load below 400 copies/ml but data were compatible with one transmission per 79 person-years. In ten studies with HIV-positive people not receiving antiretroviral therapy with 9,998 person years of follow-up, the overall HIV transmission rate, irrespective of viral load category and sexually transmitted diseases, was 5.64 per 100 person years. The largest number of serodiscordant couples was reported in five studies from Sub-Saharan Africa. "There was insufficient data to allow estimation of summary rates of transmission through sexual intercourse without condoms, or to separate female-male and male-female transmission" (Attia et al., 2009: 1399). "This systematic review did not identify any study from which the risk of HIV transmission per act of unprotected sexual intercourse among persons with suppressed viremia following ART could be quantified directly. The available studies found no episodes of HIV transmission in discordant heterosexual couples If the HIV-infected partner was treated with ART and had a viral load below 400 copies/ml…The comparison of overall rates in patients on ART and not on ART nevertheless indicated that heterosexual transmission was reduced by 92%".

An analysis of data from 463 patients (70% women) in rural Uganda on ART in a government-run clinic over 3.5 years of observation and 5,239 study visits showed that although detectable viremia and/or sexual transmission risk behavior occurred in over half of patients, ART reduced periods of HIV transmission risk by over 90% over six years of observation time. Though over half of the 463 patients had episodic detectable viremia or reported sexual transmission risk behavior, less than 5% ever experienced them simultaneously. One in three reported at least one episode of sexual transmission risk during a median of 3.5 years of observation time. Lower CD4 count, stigma, low household assets and younger age were associated with HIV transmission risk.

Using data collected at population-wide annual serological and behavioral surveys in Uganda between 1989 and 2007, no transmissions occurred in the 29 couples where the person living with HIV was on ART during 872 person years. HIV status of both partners was known in 2,465 couples and of these, 259 were HIV serodiscordant. Of the 259 serodiscordant couples, 62 converted to HIV. Higher viral load independently associated with HIV seroconversion.

A review of data from a population based prospective cohort of 16,667 individuals in South Africa who were HIV negative in 2004 and were followed until 2011 found that an HIV-negative person living in a community where 30% to 40% of those living with HIV were on ART was 38% less likely to acquire HIV than an HIV negative person living in a community where ART coverage was less than 10% after ruling out confounding factors. From 2004 to 2011, 1,413 HIV-negative people seroconverted, with a 1% increase in ART associated with a 1.4% decline in risk of acquisition of new HIV infection. Controlling for other factors, an HIV-uninfected individual was 2.2 times as likely to acquire HIV in a community where HIV prevalence was under 25% compared to under 10%. Controlling for reported condom use did not affect the strong relationship between ART coverage and the risk of HIV acquisition.  ART was delivered in 17 community-based clinics by nurses and ART counselors for those with CD4 under 200 until 2011; and from 2011, to those with CD4 under 350 in the context of a “successful, but imperfect, real-world ART program” (Tanser et al., 2013: 970). All those who tested HIV-positive were included, whether they accessed ART or not. Within a five-year period, 80% were tested for HIV. The group of 16,667 repeat testers constituted 75% of the population who were HIV-negative at the first observation. There was little correlation between community level HIV prevalence and ART coverage.

A study of 3,297 serodiscordant African couples found that each 10-fold increase in viral load was associated with a 2.9 fold increase in per-act transmission risk. The study found no viral load above which infectivity did not continue to increase.

A study reported as an abstract at the 2015 CROI found that among 234 MSM from Australia, Thailand, and Brazil, of whom 82.9% had undetectable viral load, no linked HIV transmission occurred despite close to 6,000 acts of unprotected intercourse, or 150 couple-years of follow-up.

A 2007 review of evidence for the impact of ART on sexual behavior in developing countries found three relevant studies conducted in Africa—one in Côte d'Ivoire and two in Uganda. In each study, condom use at last sexual intercourse was significantly higher among ART patients compared to non-ART patients. In the Côte d'Ivoire study of 711 patients, condom use at last sex was 80 percent for ART patients versus 59 percent for non-patients, regardless of partnership type (Moatti et al., 2003). Bateganya et al. on reported that of 926 participants in Uganda, of whom 164 received ART.  Condom use was higher among ART patients: 71 percent used condom use at last sex with a spouse for ART patients, versus 47 percent for non-ART patients (Bateganya et al., 2005). Among study participants receiving weekly home-based ART delivery and individual counseling in Uganda, Bunnell et al. found that of 723 patients, with 354 ART-naïve patients and 369 ART-experienced patients, condom use at last sex increased from 59 to 82 percent among ART-experienced patients with uninfected partners or with partners they did not previously know, and from 58 to 74 percent among ART-experienced patients with HIV-positive partners (Bunnell et al., 2006a). In individual counseling sessions, participants developed personal sexual behavior plans. Free condoms were provided. The available evidence indicates a significant reduction in risk behavior associated with ART in developing countries. However, there are few existing studies and the rigor of these studies is weak.

A study of 1,163 sexually active people in South Africa living with HIV found that ART was consistently associated with decreased sexual risk behaviors, as well as reporting a partner who was HIV negative or of unknown sero-status. Participants on ART were consistently less likely to engage in sexual risk behaviors, which held even after adjusting for HIV disease progression and time in care. Those who were on ART were less likely to have a sexual partner who was HIV-negative or of unknown serostatus.

A cross-sectional study conducted in 2005 found that consistent condom use among sexually active people living with HIV in Uganda was correlated with being on ART for more than one year. The study analyzed data on 269 sexually active ART-experienced individuals (95 males and 174 females, with 71% of men and 61% of women reporting consistent condom use after initiating ART.  Of the men, 83% and of the women, 89% believed that ART did not reduce HIV transmission. Those men and women desiring to have children were significantly less likely to use condoms. Results suggest that people living with HIV wanted to protect their HIV-negative partners from acquiring HIV and wanted to protect HIV positive partners from re-infection. Consistent condom use was less likely among those who were married, possibly due to lack of control by women over sexual decision-making within marriage, as well as a lack of understanding of the concept of serodiscordance.

An observational cohort study from 2004 to 2009 of 250 HIV-1 serodiscordant couples in Uganda found that couples reported more consistent condom use during ART use, but with no significant difference in the number of sexual partners. Of the 250 couples, 32 HIV-positive partners were started on ART.

A study of 559 HIV-positive people in Uganda, 386 women, found that ART initiation was correlated with increased condom use. Among those sexually active, unprotected sex decreased during the first 12 months from 53% to 15% and stabilized at 11.5% over the next two years.  However, believing that ART reduced the risk of HIV transmission was independently associated with higher prevalence of unprotected sex. Although women receiving ART were less sexually active than men, those women who had sex were three times more likely to report unprotected sex in multivariate analysis, possibly due to lack of availability of female condoms. Only male condoms were available at the clinic and required partner consent.

A study in South Africa with 1,544 men and 4,719 women from 2003 to 2010 from 19,703 clinics found that condom use increased following ART initiation. Of 13,604 visits at which sexual activity was reported, participants reported unprotected sex at 1,968 pre-ART visits, or 20.6% of all visits, compared to 346 post-ART visits, or 9.9% of all visits. Both men and women reported a higher frequency of unprotected sex prior to ART initiation. Unprotected sex decreased after ART initiation, with a greater decrease for men than for women. In addition, the likelihood of having more than one sex partner decreased after ART initiation ad this decrease was greater for women than for men. Decreased risk behavior was independent of CD4 cell count. 

Data analyzed from a prospective cohort of 3,381 participants living with HIV in a serodiscordant relationship in Botswana, Kenya, Rwanda, South Africa, Tanzania, Uganda, and Zambia found the proportion of visits at which reports of sex was unprotected by condoms decreased from 6.2% prior to ART initiation to 3.7% following ART initiation, with no difference between men and women, a significant difference. The number of sex acts per month did not differ prior to and following ART initiation.

A prospective cohort in Uganda of HIV-negative household members of HIV-positive patients on ART receiving home-based care found that risky sex decreased among HIV-negative adult household members. The study of 182 men and 273 women found that inconsistent condom use decreased from 29% at baseline to 15% at 24 months.

An analysis of survey data from a cross-sectional study with 85 HIV-positive women from Uganda; 50 HIV-positive women in South Africa; and 44 HIV-positive women in Brazil found that HAART users were significantly (3.6 times) more likely to use condoms. Of the 179 HIV-positive women, 83 women reporting recent sexual intercourse, with 63% using condoms and 76% using contraceptive methods. Of the 179 HIV-positive women, 65% reported currently using HAART.

A meta-analysis of randomized controlled trials of prevention interventions that measured unprotected vaginal or anal intercourse, and included skills training, group strategies for practicing safer sex, case management, positive consequences of safer sex, peer-led discussion groups, motivational enhancement, peer mentoring, and trauma coping among 11,286 people living with HIV by 2012 (21 studies) found a lower likelihood of unprotected vaginal and anal intercourse were observed in intervention arms compared to comparison arms. All studies took place in the United States. The short-term efficacy of interventions with under 10 months of follow up were statistically significant in reducing unprotected vaginal and anal intercourse. Group based interventions were more effective than individual based interventions. The effect of the intervention after ten months of follow up was not significant, suggesting that booster interventions are needed.

A cluster randomized trial of 1,891 people living with HIV on ART found that those who received a brief intervention using motivational interviewing during routine clinic care in publicly funded HIV clinical care sites in South Africa reported significantly greater reductions over a four week period in penile-vaginal or penile-anal sex without a condom with any partner regardless of serostatus. Those in the intervention group reported a 72% reduction in sexual acts without condom use from the start of the intervention until 18 months later; those in the standard of care reported 45% reduction in sexual acts without condom use, a significant difference. Those in the intervention group received one-on-one counseling with trained lay counselors assessing sexual risk behaviors; barriers to safer sex; and decided on an achievable goal. Lay counselors who were already employed in the clinics received a five-day training. The study took place from 2008 to 2010 in eight clinics randomized to the intervention and eight clinics to standard of care. Those in the intervention arm received an average of five counseling contacts, with significant increased condom use with sexual partners, regardless of their partners’ serostatus, at 6, 12, and 18-month assessments. However, no differences were found in the control and interventions groups by new sexually transmitted infections, who may or may not have received effective STI treatment. In addition, 24.6% in the standard of care group and 28.1% in the intervention group were "currently trying to have a baby."

A randomized trial of 48 women living with HIV in Nigeria who completed a six-month follow up assessment found that motivational group support resulted in significantly higher levels of condom use in the last three months, with 84.6% in the intervention group reporting condom use compared to 43.85 of those in the standard of care. Among the intervention group, 93% reported never missing any medication compared to 40% of the standard of care group. Among the intervention group, 92% of women reported not having sex because condoms were not available compared to 29.4% in the standard of care group. Facilitators received a 24-hour training. Group support focused on topics such as negotiation skills and disclosure.

An intervention with 216 couples in Zambia that addressed condom use within serodiscordant relationships to avert transmission of HIV; and within seroconcordant relationships to avert transmission of ART-resistance increased condom use over time, decreased intimate partner violence and increased positive communication. Group sessions of eight to ten participants discussed conflict resolution, sexual negotiation, effective communication and ARV adherence, as well as relaxation to respond to stress, and role-playing negotiation. All sessions were conducted separately for men and women but participants were given "couples homework" between sessions. Disclosure of serostatus was not required as part of group sessions, and group sessions included both HIV-positive and negative participants. Couples had at least six months of partnerships. Individual sessions were conducted by facilitators and provided counseling on risk reduction strategies. Condom use increased over time for both men and women who participated in group sessions, but only for men, not for women, who participated in individual sessions. Men used more positive communication strategies over time. Reports of violence decreased at 6 and 12 month follow up sessions, but only for women in individual sessions not in group sessions. Reports of use of violence predicted less male condom use. Partners who were more willing to use methods that prevented STDs reported a higher rate of condom use. Individuals who used more negative communication reported lower rates of condom use. Both individual and couples sessions increased condom use and communication between partners.

A study in Ethiopia in 2010 with 454 people living with HIV, 224 ART-naïve and 230 ART-experienced found that those who were a member of an association of people living with HIV had a 40% lower risk of using condoms inconsistently in both ART experienced and ART naïve groups. In addition, knowledge that HIV transmission can occur while on ART also was correlated with higher likelihood of condom use.

A qualitative study of 18 women with four focus groups from 10 different support groups led by para-professionals in Rwanda found that the support group empowered women to disclose their serostatus to their children, created a sense of community, gave them motivation to live, improved their self-esteem and independence, and insisting on condom use (aided by the trauma counselor) and/or choose not to have a sexual partner.

A study based on interviews with 21 male and 20 female clients in Uganda receiving ARVs from TASO in Uganda as well as ongoing counseling, found that concerns about reinfection, focusing on providing for themselves, their families and their children reduced multiple partnerships and increased condom use, but some women faced violence when requesting condom use.