Antenatal Care - Treatment

1. Initiating cART as early as possible to achieve low viral load is optimal, improves maternal health, and reduces risk of vertical transmission. [See also Pre-Conception] Note: Risk of vertical transmission rises after 28 weeks of pregnancy (7 months), so initiating ART at least by month 6 of pregnancy reduces risk (Luzuriaga and Mofenson, 2016)

A sub-study of 217 HIV-positive, pregnant women from the PACOME study, a randomized, non-blinded clinical trial conducted from December 2009-December 2011 in Benin, found that ART should be initiated prior to the start of the last trimester of pregnancy in order for a woman to achieve undetectable plasma viral load before delivery. “The longer the ART was taken, the higher the probability to achieve virologic suppression at the end of pregnancy” (para. 32). In order to be included in the study sample, women had to be living with HIV and between 16-28 weeks gestation. The observation period extended from 7 months prior to 1 year after the implementation of the new Beninese recommendations in June 2010, which changed the recommendation of starting ART from 28 weeks to 14 weeks of pregnancy. Plasma viral load was assessed twice in each woman: once at enrollment and once in late pregnancy. Self-reported adherence to ART was collected at each of 3 antenatal care visits. The primary outcome of interest was third trimester viral load, which was used as a proxy to measure efficacy and categorized as detectable (>40 copies/mL) versus undetectable (≤40 copies/mL). Most women were at an advanced stage of HIV. At the time of delivery, 71% of women had an undetectable viral load. The probability of having an undetectable viral load was more than 4-fold increased if the treatment lasted for 8 weeks or more. Three infants had a positive PCR result, and the mothers of these three infants all had viral load over 30,000 copies/mL, i.e. this showed poor health for the mothers and they were more likely to transmit HIV. Women enrolled after June 2010 were more likely to achieve virologic suppression than those enrolled prior to June 2010, which shows that starting ART at 14 weeks as opposed to 28 weeks is highly beneficial. Factors increasing the probability of obtaining virologic suppression included older age, higher weight, high antenatal care attendance (more than 6 visits during pregnancy), ART adherence, early initiation of ART, and higher CD4 cell count at enrollment.

A study of 1,684 infants enrolled in a randomized controlled trial from Brazil, South Africa, Argentina, and the United States found that higher maternal viral load was significantly associated with vertical transmission based on multivariate analysis.

A matched case control study of fifty cases and 135 controls conducted at 31 public facilities in Kenya found that women who first learned their HIV status during pregnancy were 2.85 times less likely to adhere to ART and 2.42 times more likely to have a home delivery compared to women who were on combination ART prior to pregnancy.

A hospital-based retrospective cohort study of 202 pregnant women living with HIV from 2008 to 2012 in Ethiopia found that women who initiated HAART before pregnancy (30 days before the estimated date of conception) had better immunological and clinical outcomes compared to women who initiated HAART during pregnancy. The women had a median CD4 count of 210 cells/mm3 before HAART initiation. The average duration of treatment among those that initiated HAART before pregnancy (56.4%) was 32.03 months, compared to 4.07 months among those that initiated HAART during pregnancy. Among all of the participants, 16.3% had a poor immunological outcome, defined as a decrease in CD4+ lymphocyte count between the initiation of HAART and delivery. Women with an unknown HIV status prior to pregnancy (29.7%) were 0.15 times less likely to have a good immunological outcome compared to women who knew their HIV status before pregnancy. In addition, participants with a CD4 count of less than 200 cells/mm3 before HAART initiation were 0.023 times less likely to have a good immunological outcome compared to the women who had a CD4 count greater than 200 cells/mm3. Poor immunological outcomes were also associated with women in WHO clinical stage III at baseline (32.2%). Of the 202 women, 29.7% had poor clinical outcomes, defined as a change from a lower WHO clinical stage to a higher stage between HAART initiation and delivery. Women who initiated HAART during pregnancy were 0.349 times less likely to have good clinical outcomes compared to women who initiated HAART before pregnancy. In addition, women who remained on HAART for 13 to 18 months were 0.193 less likely to have good clinical outcomes compared to those who remained on treatment for more than 18 months. Of the 2.3% of infants who acquired HIV, their mothers were on HAART for one month. Of the participants, those in WHO clinical stage III before HAART initiation were 7.673 times more likely to have a poor clinical outcome compared to women in clinical stage I. Baseline CD4 count of less than 200 cells/mm3, baseline WHO clinical stage III, and unknown HIV status prior to pregnancy were all identified as predictors of maternal treatment outcomes.

An observational cohort study, from 2000-2012 in Ukraine, of 8,884 HIV-positive mother and live-born infant pairs found that ART treatment should be started as early as possible, when an HIV-positive women’s CD4 count is still high. Majority of the women in this study (83%) started ART treatment in their third trimester of pregnancy and 54% were giving birth to their first child. A third of the women who received HIV treatment were receiving combination ART. Of the 8,884 infants born, 12% were classified as low birth weight, 9% were preterm, and 10% were small for gestational age. Preterm pregnancy was associated with injecting drug use. Analysis of a subgroup of 3,119 women, who had known CD4 counts, showed that high CD4 counts were associated with reduced risk of preterm delivery.

A prospective cohort study from 2000-2011 among 8,075 mothers living with HIV in France found that the earlier ART was started, the lower the rate of vertical transmission, whether or not a women’s viral load was less than 50 copies/mL. ART was nearly as effective when started in the first trimester as when it was started before pregnancy. In this study, there were no cases of vertical transmission among the 2,651 infants born to women who were receiving ART prior to conception and had a viral load less than 50 at delivery. The overall rate of vertical transmission was 0.7%. The rate of vertical transmission increased from 0.2% for women starting ARTs before conception, to 0.4% among those initiating in the first trimester, 0.9% among those initiating in the second trimester, and 2.2% among those initiating in the third trimester. The incidence of preterm delivery was 16% in this study, which is much higher than the general population rate. However, the risk of preterm delivery did not differ according to timing of ART initiation. The findings of this study provide strong evidence for initiating therapy as soon as possible during a women’s pregnancy.

A pre/post quasi-experimental study of pregnant women in Lilongwe, Malawi found that a greater proportion of the 14,532 women accessing ANC after the implementation of Option B+ (October 2011-March 2013) were enrolled into PMTCT services, were on ART during pregnancy, were more rapidly initiating ART, and were retained on treatment through delivery in comparison to the 13,926 women accessing ANC before the availability of Option B+ (October 2009-March 2011). The study aimed to compare service uptake and antenatal outcomes in women living with HIV pre- and post-Option B+ implementation by using routine data collected for patients enrolled in the Tingathe study. Time to ART initiation was significantly shorter post-Option B+ with 58.4% of women starting on the day of enrollment. Although more women withdrew from PMTCT services after Option B+ was available, the significant increase in enrollment with Option B+ resulted in a higher proportion of all women living with HIV receiving treatment. This study also found that those enrolled on Option B+ had higher rates of withdrawal and loss to follow up compared to women who accessed ANC before Option B+ was available. “Women may be reluctant to start ART when there is no clear indication to start for their own health, and as others have suggested, perhaps women find PMTCT coercive and so initially enroll only to withdraw later” (p. e81-e82). The study suggests that the simplification of treatment with Option B+ has resulted in several improvements in the antenatal PMTCT cascade (Kim et al., 2015). (Gray IIIb) (pregnancy, treatment, Malawi

A review from 2010 to 2011 using records from early infant diagnosis in Cameroon found that of a total of 2,505 mother infant pairs from 59 sites, found that access to Option B+ reduced vertical transmission to 4.3% compared to 31.3% among mother-infant pairs who did not receive ART.

An analysis of 12,486 infants delivered by women living with HIV from 2000 to 2011 in the U.S. and Ireland found that transmission risk was significantly lower (.09%) in women with viral loads under 50 copies/mL compared with a risk of 1% in women with viral loads of 50-399 copies/mL, regardless of ARV regimen or mode of delivery.

A study in South Africa using routine clinic records from 2010 to 2013 of 19,432 low-income women who came to ANC services found that Option B+ led to higher numbers of pregnant women living with HIV initiating ART at CD4 counts above 350. When eligibility for ART was based on CD4 counts of under 200, 18% of women who presented for ANC had initiated ART with CD4 counts under 200 when this was the cut-off for eligibility.  But under Option B+, 92% of women presenting to ANC had initiated ART prior to CD4 counts of under 200 (Myer et al., 2015). An economic assessment found it to be cost-effective.

An observational study of 311 pregnant women living with HIV in Malawi and receiving treatment under the Option B guidelines (not B+), from 2008- 2009, found that 6 of 8 (75%) of the infants who acquired HIV had mothers with a CD4 > 350, highlighting the importance of ART treatment for HIV-positive women regardless of CD4 count. ART was initiated as soon as possible among women with a CD4 < 350 and, among women with a CD4 > 350, at 25 weeks gestation or as soon as possible after they presented to a clinic. HIV transmission in this study was rare; the rate of HIV acquisition by infants was 3.2%. Half of the cases of HIV transmission were detected between 6 and 12 months postpartum, a time when almost all mothers in the study reported they had finished breastfeeding, and under Option B, were no longer on ART. A maternal baseline CD4 < 350 was the only predictor of infant mortality, which suggests that a healthier mother can improve infant survival. Almost one-third (30%) of women who discontinued treatment at 6 months postpartum, under the Option B guidelines, had to resume it by 24 months, either because of a new pregnancy or low CD4 count. One in five women reached the CD4 count criteria for treatment by 18 months of discontinuation of ARTs. The probability of loss to follow up was 16.4% at 2 years, and a CD4 count > 350 was a predictor, which suggests that Option B may have a negative effect on women returning to care. In this study, effective treatment initiated in the last trimester of pregnancy was not sufficient to reduce the risk of vertical transmission.

A retrospective, observational cohort study of 10,150 pregnancies, from 2002-2010, among 8,661 women living with HIV in Malawi and Mozambique found that mortality among women who received triple ART for less than 30 days prior to delivery was 3-fold that of women who received triple ART for 3 months or more prior to delivery. Data from pregnant women living with HIV who attended any of the 16 Drug Resource Enhancement Against AIDs and Malnutrition Program (DREAM) study centers for prenatal care were evaluated in this study. A total of 8,172 women initiated triple ART during prenatal care, while 1,978 women were already on triple ART prior to pregnancy. Short-term mortality, defined as death of the mother during pregnancy or within 42 days after delivery, was significantly reduced with longer duration of antenatal triple ART. Women who received triple ART for the shortest duration, 0-30 days prior to delivery, had the highest mortality rate at 2.2%. Women who initiated triple ART 31-90 days prior to delivery had a lower mortality rate of 1.1%. However, the women who were on triple ART treatment for at least 3 months before delivery had the lowest mortality rate of 0.6%. The study also found that the major factors associated with long-term maternal mortality for women living with HIV, defined as death of the mother between 42 days to four years after delivery, were less than 30 days of triple ART before delivery. In this study, women on ART before pregnancy did not have a higher mortality rate than women initiating ART during prenatal care, despite having a more advanced disease with lower CD4 count, which shows the significant benefit of women initiating ART for their own health, rather than just during pregnancy. The DREAM study has routinely provided HAART to all women living with HIV for PMTCT in several sub-Saharan African countries since 2002, irrespective of CD4 cell count, during prenatal care and breastfeeding.

Retrospective data from 2,692 mother-infant pairs in Canada from 1997, when cART became the standard of care, until 2010, found that the rate of transmission for mothers who received cART was 0.4% if the mother living with HIV had more than four weeks of cART. The transmission rate was 9% for women who received less than four weeks of cART.

A prospective cohort study, which followed 1,393 ART naïve pregnant women living with HIV, from 2003-2007 in Cameroon, Côte d’Ivoire, Kenya, Mozambique, Rwanda, South Africa, Uganda, Zambia, and Thailand found that 1 in 4 women who received ARV prophylaxis during pregnancy and were discontinued from treatment at delivery became eligible for ART, under Option B, within 24 months of delivery. All 1,393 women in the study had a CD4 count ≥ 250 and 903 had CD4 count ≥ 400. Majority of the women were on either single-dose nevirapine (45%) or short course ART prophylaxis (38.2%) while just 12.3% received triple ARV. By 2004, sites in Bangkok, Thailand and Eldoret, Kenya offered triple ARV prophylaxis. All women were taken off treatment at delivery, as per WHO guidelines at the time, and CD4 count decline was measured at 12 and 24 months postpartum. Among women who had a CD4 count ≥ 250 at enrollment, 4.5% had declined to CD4 < 200 at 12 months and 11.6% had declined to CD4 < 200 at 24 months. Among women who had a CD4 count ≥ 400 at enrollment, 11.9% had a CD4 < 350 at 12 months and 27.5% had declined to CD4 < 200 by 24 months. Women who received triple ART had a higher probability of CD4 decline by 24 months compared to those receiving other antiretroviral regimens, which suggests that triple ART interruption may be associated with a more rapid decline than other ART regimens. The majority of women (60.3%) on triple ART who had CD4 of 400-499 at enrollment declined to <350 by 24 months postpartum. After discontinuation of triple ART, women lost on average 20 CD4 cells/mm3 per week in the first 8 weeks and 2 CD4 cells/mm3 per week thereafter. Higher CD4 cell count at enrollment was associated with a reduced probability of immunological decline; for each increase in 100 cells/mm3, the probability of reaching CD4 < 200 was reduced by 40%. This study shows that a substantial proportion of women who receive ARTs during pregnancy will rapidly decline in CD4 cell count within 24 months if discontinued from treatment after delivery.

A 2007-2010 retrospective cohort study in Zambia analyzed data on 1,813 HIV-positive pregnant women attending antenatal clinics to assess various exposures of mother-to-child-transmission. The study found that the odds of vertical transmission increased 5.5-fold among women on HAART for 4 weeks or less before delivery, compared to those on HAART for 13 weeks or more. For each additional week on HAART (up to 13 weeks) before delivery, the odds of transmission were reduced by 14%. In this cohort, mother-to-child-transmission of HIV occurred in 3.3% of infants (59 in 1813). Mother-infant pairs were considered eligible for this study if mothers began HAART during pregnancy and if their infants had an HIV test result assessed by PCR from 3-12 weeks of age. Infant HIV status was the primary outcome. Electronic records provided comprehensive mother and newborn data through the first six weeks, which included HAART initiation, gestational age, demographic characteristics, infant birth weight and CD4 cell count. HAART duration was categorized as 4 weeks or less, 5-8 weeks, 9-12 weeks or 13 weeks or more. Maternal age, infant weight at birth, maternal BMI or hemoglobin levels, maternal CD4 count and gestational age were not found to be associated with infant HIV infection by 12 weeks.

A retrospective cohort study from 2004-2008 followed 418 HIV-positive mothers and their infants who participated in a routine PMTCT program in Cameroon. The study found that ART regimens lasting less than four weeks during pregnancy led to a 4.7-fold higher risk of early vertical transmission prior to ten weeks of age for the infant. 

A study from Malawi analyzing the national evaluation of the Option B+ program found that women on ART prior to pregnancy had low rates of vertical transmission.

Preliminary results from a study in Swaziland found that Option B+ resulted in more women initiating ART (94% of 1043 women) than on Option A (35% of 1272 women). 

Back to Top

2. Peer counseling by mother mentors may improve treatment adherence among pregnant women living with HIV.

A cluster randomized controlled trial in South Africa found that peer mentors supporting women living with HIV and their infants resulted in significantly fewer depressive symptoms and fewer underweight babies, as well as greater adherence to the guidance at the time on prevention of vertical transmission. The women who received the internvetion were 1.08 times more likely to not be depressed than those receiving the standard of care (p = 0.002). Eight clinics were randomized for pregnant women living with HIV to either receive standard of care or an intervention with peer mentors of women living with HIV who had received training. There were eight meetings that discussed establishing healthy routines; adhering to ART; couple disclosure; consistent condom use; and infant bonding. Peer mentors were trained, and had weekly supervision. After twelve months, outcomes for 181 women in the standard of care and 106 women who attended at least one session with peer mentors was analyzed.

An evaluation of a mentoring program in health facilities of mothers living with HIV in Uganda found a statistically significant increase in retention in care of women living with HIV on triple ART (90.9%) compared to health facilities without this support (63.6%). Additionally, there was a statistically significant reduced rate of vertical transmission, from 6.8% in facilities with mentor mothers, compared to 8.7% where there were no mentor mothers. The study was conducted in 31 health facilities with mentor mothers and 32 health facilities without mentor mothers. A total of 1,150 mother and baby pairs who received services between 2011 and 2014 evaluated. Pregnant women living with HIV who attended health facilities with mentor mothers were more likely to have the WHO-recommended four antenatal care visits during pregnancy and give birth with a skilled birth attendant. A survey of 400 pregnant women in facilities with and without mentor mothers from 2012 to 2014 found that women living with HIV reported statistically significant improved psychosocial wellbeing in facilities with mentor mothers, with reported improved coping self-efficacy, safer sex, and reduced levels of depression. The study found that mentor mothers were a cost-effective intervention, with US$1 spent on mentor mothers averting $11.40 in associated treatment costs.

A review of 21,939 patients in Malawi who started ART under Option B+ at one of 540 facilities from 2011 to 2012 in Malawi found that facilities that offered additional adherence counseling as required by national guidelines, which included peer counseling by mother mentors, had lower rates of early loss to follow up.

A cross-sectional study of adherence among 277 pregnant women living with HIV in 2014 in Ethiopia found that the odds of adhering to Option B+ were 4.7 times higher among women who received counseling on the importance of adherence to ART from mother support groups and clinic staff.

Focus group discussions with 106 women living with HIV who had given birth in the prior three years from Nigeria, Kenya and Namibia found that women reported that the information from mentor mothers was more useful than information than that received by health providers and led to more needed support.

A qualitative study found in Uganda found that peer educators may increase retention of pregnant women living with HIV, increasing adherence and support (ICW et al., 2016).  

Back to Top

3. Community health workers and community-based support can increase uptake of safe motherhood interventions for women living with HIV and reduce vertical transmission

A cluster randomized controlled trial of women living with HIV in South Africa found that 644 women who received home visits by community health workers were significantly more likely to avoid birth related complications, take actions to reduce vertical transmission and have healthier infants, as well as use condoms more consistently, in comparison to 169 women who received standard of care at clinics.  The odds of completing all the tasks to reduce vertical transmission were 1.95 higher among those women visited by community health workers. The tasks to reduce vertical transmission were standard of care at the time, such as administering Nevirapine at birth to the infant, correctly medicate infants and breastfeed exclusively for the first six months. Mothers and infants were followed for six months postpartum. Community health workers were selected for good communication skills and were trained for one month using role-playing and key health information. Community health workers were paid US $150 per month. On average, community health workers made six antenatal visits and five postnatal visits per woman, averaging 31 minutes per visit. “By having community health workers identified with a maternal, child health and nutrition program, much of the stigma associated with HIV is side-stepped”.

A pilot program in South Africa with 50 pregnant women living with HIV who had support from case managers and text messages were statistically significantly more likely (90% vs. 63%) to have had their infants tested for HIV postpartum than a comparison group of 50 pregnant women living with HIV. All women had a cell phone. In the intervention group, case managers who were lay counselors, sent a pre-scripted text message until six weeks post-partum. Case managers also made a phone call prior to delivery and two phone calls postpartum. Women could request a phone call at no charge from the case manager.  The program cost US$364 in cell communication over four months plus US$29 per cell phone for the case manager. Women found the intervention acceptable and that it provided needed emotional support as well as a resource to ask questions. Messages included, “A healthy baby starts with a healthy mother! Be sure to take your tablets every day. ...Congratulations on your new baby. I hope this is a special time for you”.

A study in Kenya where community health workers tracked 650 pregnant women living with HIV through a mobile health tool using text messages to remind women of their appointments found a significantly lower rate of vertical transmission (0%) than a comparison group (9%) and women were twice as likely to attend more antenatal care visits. Women who missed appointments were visited at their home by community health workers, but the study did not assess if the women felt this was too invasive, violated their privacy or resulted in any adverse outcomes, although the texts did not refer to HIV serostatus. In addition, not all women have access to a phone or their phone is shared with others in their household.

An evaluation assessed the effect of HIV programs supported by PEPFAR on maternal health services in 257 facilities in eight African countries from 2007-2011 found that facilities that offered support groups for women living with HIV had 6% more deliveries at the facility than facilities that did not have support groups.

A cohort study of 1,105 pregnant women living with HIV in South Africa from 2009 to 2012 found that the rate of ART initiation was 57% greater among the 264 women who received a community- based support intervention. Community-based support healthcare workers visited pregnant women at their homes to provide HIV education and ART counseling, promote PMTCT, and address various psychosocial barriers to ART, including lack of partner involvement, non-disclosure, stigmas, fear of ART, nutrition insecurity, depression, gender based violence and social assistance grant eligibility. ART- eligible women were visited by the community-based support healthcare workers three times during the ART initiation week, then visited weekly for a month, and then visited once a month. The median baseline CD4 cell count was 305 cells/μL among women who received the community based support intervention and 361 cells/μL among women who did not receive the intervention.  Among all participants eligible to initiate triple ART, 5.4% of those that received the community based support intervention did not initiate ART, compared to 30.3% among those that did not receive the intervention. Participants who received community-based support were more likely to initiate ART with less delay, with a median of 26 days compared to 39 days among those that did not receive the intervention. In addition, ART coverage among all women at delivery was 64.8% among those that received community based support and 38.5% among those that did not.

A project from 2011 to 2013 in Malawi to increase use of Option B+, i.e. treatment for life, that worked in five districts found that women’s community-based support groups, was positively correlated with increased maternal use of ART as measured with cross sectional data. Additional interventions included health working training and mentorship, improved lab systems and couples testing and counseling.  The study results did not include those pregnant women living with HIV who did not have at least one ANC visit. Prior to Option B+ cART uptake was 23% among women living with HIV; following implementation of the program, ART uptake increased significantly to 96% by 2013. However, access to transport or funds for transport remains an issue for pregnant women living with HIV.

A pilot program in Malawi using community health workers (CHW) resulted in increased uptake of ART as per guidelines of the time, i.e. ART eligibility at CD4 counts under 250. Before the intervention, only 8.8% of pregnant women received ART as per eligibility; following the intervention, 40% of pregnant women living with HIV received ART per eligibility. Over 24 months, 1,688 pregnant women living with HIV were enrolled. Of 499 women eligible for ART, 72.8% were initiated on ART. Prior to giving birth, 1,264 women received ARV prophylaxis. Community health workers were tasked with providing community education and ensuring that pregnant women living with HIV were on ART. The two-week CHW training included adherence counseling and reducing stigma. CHWs tracked women at clinics and at their homes from initial HIV diagnosis until cessation of breastfeeding or ART initiation for infants with HIV, verifying that women ingested ART. NOTE:While the study did not address this, this may have been seen as coercive by some women: “We also need to prevent women from refusing care and dropping out, for whatever reason”.

A nationally representative household survey in 2009 in Rwanda of mothers who attended ANC at least once during their most recent pregnancy found that HIV-free survival of HIV-exposed children was correlated with being a member of an association of people living with HIV, after adjusting for maternal, child and health system factors.  HIV exposed children born to mothers living with HIV were tested. Of 1,448 HIV-exposed children surveyed, 4% were reported dead by nine moths of age. Of 1,340 children alive, 4% tested HIV-positive. HIV-free survival of HIV-exposed children was 91.9%.

A randomized study of 10 PMTCT sites between 2013 and 2015 in the Maternal-Infant Retention for Health study in Kenya found that women who received lay counselor home visits and support had a significantly lower loss to follow up six months post partum compared to standard of care. Lay counselors provided individual health education at clinics and during home visits, appointment reminders, physical tracing after a missed clinic appointment and individual adherence support to 170 women and compared this to 170 women who did not have lay counselor support. At six months post partum, 130 mother-infant pairs remained in care with lay counselors compared to 112 in the standard of care. Loss to follow up was significantly lower among women who had lay counselors - 18.8% compared to 28.2% in those with standard of care.

Back to Top

4. PMTCT-Plus (family-focused) HIV care can increase the numbers of women and their partners who access treatment and remain adherent. Note: This should not be implemented in any way that prejudices women who do not want to disclose to partners (see overview).

A study of 4,278 adults (3,613 women) enrolled in HIV care through the MTCT-Plus Initiative from 2003 to 2008 in 11 African countries found that women with other family members enrolled in the program had the lowest loss to follow-up rates over the course of the study (16.7%). Among women, 8.7% of the lost to follow up rates were attributed to not having a family member co-enrolled. The program was family-focused and provided comprehensive HIV care, including ART for eligible participants (according to national or WHO guidelines), as well as physical examinations, and CD4 cell counts. All participants’ partners and family members living with HIV were also eligible for the program. The median age of participants was 27 years of age among women and 33 years of age among men. All participants had a CD4 count greater than or equal to 200 cells/ mm3 (median of 441 cells/ mm3), were in WHO clinical stage I or II, and had at least one follow up visit after the initial enrollment visit. Of the women, 46.4% were pregnant at baseline and 55.2% were pregnant at some point during the study period. At 12 months follow up, the median loss to follow up (12 months since last visit) was 8.2%, ranging from 0.3% in Uganda to 21.8% in Kenya. At 24 months, median loss to follow up was 15.6%, ranging from 1.5% in Uganda to 35.9% in Kenya. This number was higher among women, with a median of 8.4% and 16% at 12 and 24 months follow up, compared to 7.1% and 13.5% among men. At the end of the study, 26.4% of the original sample were lost to follow up, with 1.4% known to have died before initiating ART, 30.2% had initiated ART, 8.1% were documented to have withdrawn from the program, and 38.2% had never initiated ART but were alive and in care. Across both genders, younger participants were found to be at a higher risk of loss to follow up, with those between 15 and 24 years of age at a higher risk than those over 30 years old. In addition, women with a higher CD4 count (350-500 cells/ mm3) were 1.5 times more likely to be lost to follow up than those with a CD4 count less than 350 cells/ mm3. Being pregnant at a previous clinic visit was also associated with a higher risk of being lost to follow up, with 33.2% of loss to follow up among women attributed to pregnancy. Among all participants, living in a household with four or more people was found to decrease the risk of becoming lost to follow up. In addition, men with electricity in their home and women with employment outside of their home were less likely to be lost to follow up.

A study in Northern Uganda from 2002-2011 analyzed results from 24 health facilities in Northern Uganda with 140,658 women who attended ANC. The increased ANC attendance in the first few years “paralleled increased access to, and sustained sensitization about the availability of PMTCT services” (p. e143). Male partner attendance increased from 5.9% in 2002 to 75.8% in 2011. HIV-prevalence in HIV-exposed infants decreased from 10.3% in 2002 to 5.0% in 2011. The Uganda Ministry of Health program created Family Support Groups, which were implemented to promote community support to HIV sero-discordant and concordant couples and their infants. These groups discussed health education and implemented social support and income-generating activities in collaboration with the health facility. Men had also been included through peer counseling by other men and the creation of male-friendly spaces in ANC clinics. Even though conflict was prevalent in the region until 2006 these PMTCT program outcomes were comparable or better than non-conflict areas. Therefore, “a comprehensive PMTCT program emphasizing social and community engagement alongside medical care and support can succeed in a remote setting with multiple challenges” (p. e138). The Ugandan Ministry of Health increased antenatal care and decreased rates of vertical transmission.

A systematic review that included 20 articles that met the inclusion criteria found that providing family focused care increased women’s uptake of HIV-related services. Studies took place in South Africa, Kenya, Tanzania, Zambia, Botswana, and Côte d’Ivoire.

A study from Côte d’Ivoire evaluating an MTCT-plus program from 2003 to 2005 found a significant increase in antiretroviral treatment initiation and high rates of retention in care for women and their partners. Of the 605 women enrolled during the study period, fewer than 2% of women and 9% of their partners were receiving antiretroviral treatment prior to enrollment in the program, in comparison to 41.5% of women and 65% of their partners after enrollment at the close of the study period. Retention rates were also high: only 2.5% of women and 5.5% of partners initiating ART were lost to follow-up, while 2% of women and 0% of partners not eligible for ART were lost to follow-up.

Back to Top

5. Integrating ARV therapy into antenatal care, rather than referring women separately for HIV treatment, can reduce time to treatment and increase adherence for pregnant women living with HIV. [See also Structuring Health Services to Meet Women’s Needs] Note: A review found that no one model of care fully addressed all barriers for women, but the most effective models focused on the period of transition between pregnancy and postpartum (Colvin et al., 2014).

A quasi-experimental nonrandomized study in Zambia from 2011 to 2013 found that the average time to ART initiation among ART-naïve pregnant women living with HIV who attended ANC and HIV integrated clinics was 22.2 days, compared to 31.8 days among ART- naïve pregnant women living with HIV who attended standard of care ANC services. The intervention included training 132 ANC providers in ART initiation and management. Participants in the study were assigned to lay counselors who made regular home visits throughout pregnancy and breastfeeding to assist with navigating the health system, promote adherence, and provide patient support.

A Cochrane review with three studies on integrating ANC and ART found that the proportion of pregnant women initiating ART during pregnancy increased by 32.9% when ANC and ART services were integrated, compared with those pregnant women who were referred to HIV services. In addition, interventions that integrated ART and ANC reduced the delay between HIV diagnosis and initiation of ART from 56 days to 37 days (Lindgren et al., 2012 cited in Lisy, 2013). 

A review of 279 postpartum women in South Africa who started ART in the ANC clinic who were transferred to ART clinics postpartum between 2012 and 2013 found that those who had additional months on ART before delivery had a greater likelihood of engagement in an ART clinic postpartum. Of the women transferred, 32% were transferred to a large ART facility on the same premises as the ANC clinic.  After adjusting for age, CD4 cell count and being diagnosed with HIV in the current pregnancy, the relative risk of successfully engaging in care increased by 5% for every additional month on ART before delivery. Based on only lab assessments, 74% were engaged in care after transfer.  However, 91% of women self-reported engagement in care postpartum.

A study in South Africa using routine clinic records from 2010 to 2013 of 19,432 low-income women who came to ANC services found that service integration of ART with ANC led to high rates of ART initiation as compared a system in earlier years which required referral between ANC services and ART treatment services.  Compared with the model with ART eligibility based on CD4 counts under 350, women were approximately seven times more likely to initiate ART in the services that integrated ART with ANC and more than 20 times as likely to initiate ART before delivery under Option B+. Of pregnant women in this community, more than 95% attend ANC prior to delivery. Six service delivery models were evaluated in sequence: 1) ANC services referred all ART- eligible women based on 2006 WHO guidelines to the ART clinic or CD4 counts under 200; 2) ANC referred all ART eligible women based on 2006 WHO guidelines to the ART clinic or CD4 counts under 350; 3) Lay PMTCT counselors worked as patient navigators to support referrals between ANC and ART; 4) ART-trained midwives initiated ART within ANC services; 5) ART-trained midwives initiated ART within ANC services with on site CD4 testing; and 5) Option B+, with CD4 counts used only to determine baseline CD4 but not for ART eligibility and ART provided at ANC if a woman tested positive for HIV.

A prospective cohort study of 321 pregnant women living with HIV attending prevention of mother-to-child transmission services from 2011 to 2014 in Zambia found that women who attended referral health facilities were more likely to be non-adherent to ART, compared to women who attended non-referral facilities. Women in the study visited 11 health facilities, six of which provided HIV care and treatment and five of which referred antiretroviral eligible women to other health centers that could provide treatment. Of the participants, 48% were already receiving antiretroviral therapy before participating in the study, and 49.5% were newly diagnosed as living with HIV during their current pregnancy.

A retrospective cohort study using routinely collected clinic data in 2008 among 14,617 women in South Africa seeking ANC, of whom 30% were living with HIV and 17% had CD4 counts under 200, found that a significantly higher proportion (55% compared to 45%) of women started ART during ANC in integrated care than when women were referred from ANC for ART services.

A study in Mozambique found that integration of HIV/AIDS services into ANC services reduced loss to follow up of women living with HIV from PMTCT services to ART services by 70% compared to 25% achieved in vertical sites. The study assessed the changes between 2004 and 2008, when HIV care was delivered through a vertical hospital and HIV care was integrated into primary healthcare. In 2005, only 30% of pregnant women who tested HIV-positive in PMTCT programs enrolled in HIV treatment and care. By the end of 2005, only 20% of eligible mothers had initiated ART In 2004, freestanding HIV treatment hospitals were constructed in urban centers with their own pharmacies, data systems, health workforce, waiting areas and receptions. Patients identified as HIV-positive from other sectors of the health system, such as PMTCT or HTC, were referred to HIV hospitals to register for HIV care.  But in 2005, only 78% of HIV-positive patients referred to HIV hospitals returned for CD4 testing, and only 46% of those who returned for the results of their CD4 tests were found to be eligible to start antiretroviral therapy.

The International Center for AIDS Care and Treatment Programs (ICAP) collected program data from 32 antenatal clinics in Rwanda from 2006-2008, where 2,048 pregnant women living with HIV attended either standard PMTCT sites (where pregnant women were referred to ART clinics that were off-site) or integrated sites, where all services for pregnant women living with HIV were provided at the same clinic, including antiretroviral therapy.  The study found that women attending integrated sites were 30% more likely to undergo CD4 cell count testing during pregnancy and twice as likely to enroll in antiretroviral treatment compared to women attending standard sites, where they were referred for antiretroviral treatment. Scale up between 2006 and 2008 resulting in increased CD4 cell count screening during pregnancy increasing from 60% to 70% and initiation of HAART from 35.5% to 97%. No differences were observed regarding HAART initiation for women determined to be eligible (about 85% in both sites) and type of treatment provided, indicating effective referral from standard sites to antiretroviral treatment services. Women were eligible for HAART with a CD4 cell count below 350. About 24% of women were eligible for HAART and 83% initiated HAART during pregnancy, regardless of service delivery (integrated or standard). Both sites provided dual antiretroviral and single-dose nevirapine regimens, while integrated sites also offered HAART and HAART to prevent vertical transmission during pregnancy until delivery or until the end of breastfeeding. Corrective strategies for scale up included providing CD4 machines and trained staff at the district level; with scheduled weekly CD4 sample processing and home visits conducted to track women who missed appointments. Most standard sites did not provide CD4 testing and referred eligible women to local ART centers for testing and treatment. The study trained and retrained 297 staff to administer multi-drug antiretroviral therapy and provided regular on-site mentoring. Study sites provided monthly reports.

A qualitative study in rural Kenya found that women living with HIV preferred integrated ANC and HIV services because this way they would not be as easily identifiable as living with HIV, as well as have easier access to comprehensive services.

Back to Top

6. National scale-up of cART in pregnancy improves maternal and infant outcomes.

A study in South Africa in 2013 found that a national scale-up of prevention of vertical transmission services significantly increased prevention of vertical transmission and treatment of women living with HIV. The overall vertical transmission rate was reported at 3.5% in 2010 and dropped to 2.7% in 2011. South Africa did not implement their first prevention of vertical transmission program until 2002. In 2004, South Africa then moved to comprehensive care management and treatment of all people, including pregnant women, living with HIV. In 2008, the Ministry of Health launched the accelerated prevention of vertical transmission program with a goal of reducing vertical transmission to less than 5% by 2011. In 2010, 30.2% of all pregnant women who sought care at public sector health facilities were living with HIV and by 2011, “70.4% of all maternal deaths in South Africa were associated with HIV infection…” (p. 70). In 2010, the Ministry of Health revised the vertical transmission policy to align with the new WHO recommendations to include lifelong HAART for women living with HIV with CD4+ counts less than 350 cells/mm3. A major effort from 2008-2011 shifted practices toward nurses initiating and managing the use of ART by training them to provide ART for all pregnant women at primary care ANC facilities. In 2005, fewer than 50% of all pregnant women were routinely tested for HIV but by 2009 the testing was “virtually universal” (p. 71). Testing of infants to detect infection before 2 months has increased from 36.6% in 2008 to 70.4% in 2011. In addition, the proportion of infants tested who were HIV-positive decreased from 9.6% to 2.5% over the same period. Rapid implementation of changes in PMTCT policy was key to scale up as well as challenging the lack of action by the government until 2001 in court, winning the court case and forcing the government to scale up services.

An observational cohort study, from 2000-2012 in Ukraine, of 8,884 HIV-positive mother and live-born infant pairs found that it is important to ensure continuing efforts to improve pregnancy outcomes among women living with HIV, and that some risk factors for adverse pregnancy outcomes are specific to HIV while others are shared with the general antenatal population. The majority of the women in this study (83%) started ART treatment in their third trimester of pregnancy and 54% were giving birth to their first child. A third of the women who received HIV treatment were receiving combination ART. Of the 8,884 infants born, 12% were classified as low birth weight (< 2,500 g.), 9% were preterm (< 37 gestational weeks), and 10% were small for gestational age (lower than the 10th percentile). The proportion of women who received no antenatal care declined significantly from 78% in 2000, to 52% in 2001, to 9% in 2012. Furthermore, there was a dramatic shift towards starting ART earlier over the time period of the study. Median gestational age at ART initiation was 34 weeks before 2005, 28 weeks in 2005-08, and 24 weeks in 2009-2012.

A study conducted in Malawi from July 2011 to September 2012 analyzed the introduction of the Option B+ prevention of vertical transmission strategy. Because obtaining CD4 counts to assess eligibility for ART can be a barrier, Option B+, which only requires an HIV-positive diagnosis in pregnancy, can increase the numbers of pregnant women who access ART. In Malawi, the number of pregnant and breastfeeding women started on ART per quarter increased by 748% from 1,257 in the second quarter of 2011 (before Option B+ implementation) to 10,633 in the third quarter of 2012 (one year after implementation). Of the 2,949 women who started ART under Option B+ in the third quarter of 2012 and did not transfer care, 2,267 (77%) continued to receive ART at 12 months; a retention rate similar to the rate for all adults in Malawi. To decentralize ART to all health centers providing ANC, 4,839 health-care workers were trained. The total number of all persons started on ART per quarter increased by 61% after implementation of Option B+. The provision of ART at all ANC health centers reduced the transportation and cost barriers to receiving treatment. Members of a nationally coordinated supervision team visited every integrated prevention of vertical transmission and ART site in Malawi quarterly. Patient registers were created to permit longitudinal follow-up and cohort analyses for patients receiving antenatal and HIV care.

A study in Jamaica in 2012 analyzed the successes and challenges of the prevention, treatment, and care of pediatric, perinatal, and adolescent HIV/AIDS in Jamaica. It concluded that Jamaica had achieved its goal of less than 2% vertical HIV transmission rate as well as more than 95% of mothers attending ANC tested for HIV. The vertical transmission rate in 2005 was 10% whereas in 2011 it was only 1.19%. The percent of women receiving ARVs increased from 74% in 2005 to 85% in 2011 while the percent of infants receiving medication to prevent vertical transmission increased from 87% in 2005 to 101% in 2011. The incidence of vertical transmission has dropped to 0.3 cases per 1000 live births. In addition, the number of HIV-exposed infants decreased from 407 in 2005 to 350 in 2011. The use of HAART for pregnant women along with a comprehensive system of care has “greatly decreased HIV/AIDS attributable maternal morbidity and mortality” (p. 398). More than 85% of women received ARTs and 100% of babies received ART chemoprophylaxis.

Back to Top