South Africa is moving towards an AIDS-free generation, and improving survival in children living with HIV

This article originally appeared in HIV & AIDS treatment in practice, an email newsletter for healthcare workers and community-based organisations in resource-limited settings published by NAM between 2003 and 2014.
This article is more than 13 years old. Click here for more recent articles on this topic

Reductions in vertical HIV transmission to infants and improvement in paediatric survival attributable to the scale of and quality improvement of prevention of parent-to-child transmission (PPTCT) and antiretroviral therapy (ART) services for children were some of the biggest success stories reported at the South African AIDS Conference in Durban this June. Many HIV-positive children are living into adolescence, however there are also worrying signs that treatment options currently available in the country may not be adequate to keep all of them in good health into adulthood.

Leading off the good news were reports that South Africa is dramatically reducing perinatal HIV transmission among HIV-exposed infants. HIV prevalence is down to 3.5% at the first immunisation visit (4-8 weeks post-partum), according to the first national surveillance study of PPTCT (still often called prevention of mother-to-child transmission (PMTCT), conducted in the last half of 2010.1 These findings were met with nothing short of jubilation — one report suggested that over 67,000 infants had been saved from HIV infection.

However, Dr Ameena Goga of South Africa’s Medical Research Council, who presented the survey findings, cautioned that the percentage of children who ultimately become infected will be higher, particularly since a significant proportion of the children are not being fed safely (with either exclusive breastfeeding or exclusive formula feeding) by their caregivers (see What about late postnatal HIV transmission? below).

Glossary

vertical transmission

Transmission of an infection from mother-to-baby, during pregnancy, childbirth, or breastfeeding.

 

immunisation

Immunisation is the process whereby a person is made immune or resistant to an infectious disease, typically by the administration of a vaccine. Vaccines stimulate the body’s own immune system to protect the person against subsequent infection or disease.

 

antenatal

The period of time from conception up to birth.

polymerase chain reaction (PCR)

A method of amplifying fragments of genetic material so that they can be detected. Some viral load tests are based on this method.

referral

A healthcare professional’s recommendation that a person sees another medical specialist or service.

Nevertheless, poster presentations provided preliminary indications that much, though not all of the prevention benefit, is being sustained over time in some sites with well-supported programmes. While the better outcomes are partly a result of improved PPTCT guidelines (see below) the poster presentations also provided a wealth of information about a number of specific innovations that have been implemented at some sites, and that could be scaled up more widely to improve service delivery of some of the key PPTCT interventions.

Background on prevention of vertical transmission in South Africa

Since the first projects to prevent vertical HIV transmission to infants were launched in South Africa in 2001 — after a hard-fought legal battle with the government — there have been several isolated reports on the impact of the intervention in the field. For most of the last decade single-dose nevirapine (sdNVP) at labour was the key intervention offered to pregnant women who had tested HIV-positive.

For instance, Dr Goga noted that the earliest report, a record review at Helen Joseph/Rahima Moosa Hospital from 2001-2002, reported an 8.7% HIV transmission rate in exposed infants at six weeks post partum. Similarly an assessment of the programme in Khayelitsha in the latter months of 2003, which used AZT prophylaxis from week 34 of pregnancy, reported that a similar percentage (8.8%) of exposed infants were HIV-infected when tested at week 6-10 post-partum.

Subsequently, the Good Start Study, conducted from October 2002 to November 2004 at three sites (Paarl (Western Cape), Umlazi (KwaZulu-Natal), and Umzimkulu, (Eastern Cape)), reported highly variable programme performance.2 Perinatal HIV transmission ranged from 8.6% in Paarl up to 13.7% in Umzimkulu — but the study also found that much of the benefit of sdNVP was lost over time due to ongoing exposure to HIV in breastmilk, especially in Umzimkulu, where HIV-free survival at 36 weeks was only 65%. Poorer outcomes were partly due to feeding practices, as well as factors such as maternal viral load, prematurity, socioeconomic score, access to antenatal care and the quality of counselling received.

 Distressing as these findings were, a subsequent study that measured both programme reach and effectiveness found very high rates of infection, 20.2%, among HIV-exposed infants attending their first immunisation at seven clinics in KZN (between Aug 2004 to July 2005).3 There was no improvement in a subsequent study at three primary health clinics in KZN between November 2007 and February 2008, with 21.9% of HIV-exposed infants testing positive.4

 In some parts of the country at least, the programme was clearly failing to adequately protect HIV-exposed infants.

Consequently, reducing vertical transmission became one of the highest priorities in the South African National HIV/AIDS Strategic Plan (2007-2011). The goal has been to reduce HIV transmission to 5% in HIV-exposed infants by 2011. To accomplish this, performance would have to be improved along each step of what has been called the PMTCT cascade — meaning the separate interventions involved in the programme including coverage, HIV testing uptake, results delivery, intervention delivery, follow-up and support, etc.

In addition, two major updates have been made to the national regimen for prevention of vertical transmission (PVT), adding further steps to the cascade. First, in 2008, all HIV-positive pregnant women with CD4 counts below 200 were to be put on ART (for both their own health and to reduce transmission), while those with CD4 counts above 200 were to be placed on AZT from 28 weeks of pregnancy and given sd-NVP at labour. The infant was also to be given sd-NVP at birth followed by AZT to day 7 and then given an HIV DNA PCR test, 4-6 weeks after delivery (not just to monitor programme performance but because identifying HIV-infected infants and putting them on ART quickly dramatically improves survival). This new regimen would require much better linkage and/or integration between the PVT programme, the ART programme in order to provide CD4 cell testing and ART, as well as maternal child health services.

Then in April 2010, South Africa updated its guidelines to offer ART to all pregnant women with CD4 cell counts below 350, and put the rest on AZT from week 14 after gestation, dropping sd-NVP for the mother as there was a potential for the development of resistance that could limit her future treatment options. Instead, the infant was to be placed on nevirapine for at least 6 weeks or for the duration of breastfeeding. Again early infant diagnosis was to be performed at the first immunisation visit.

Although there was no structured surveillance to evaluate the operational effectiveness of the programme, data from the National Health Laboratory Service submitted from infants under two months of age was 8.2% in 2008, dropping to 5.8% in 2009. However, these data could be biased because mothers who were most adherent to the programme would be most likely to bring in their children for HIV testing. A more rigorous study was needed to monitor the programme’s effectiveness.

The South African PMTCT Evaluation (SAPMTCTE)

The South African PMTCT Evaluation is a cross-sectional survey designed to periodically assess HIV transmission in infants at their first immunisation visit. In addition, the survey was designed to assess coverage of the key interventions and services for prevention of vertical transmission, and look for any associations between the vertical transmission and ART regimen, maternal background characteristics, healthcare services and maternal and infant health status.

The study included infants aged 4 to 8 weeks making their first immunisation visit (excluding older or younger infants, or infants requiring emergency care) at randomly selected primary and community healthcare clinics (n=580) in each of the nine provinces (note, a limitation of this study is that it did not include hospitals). The study aimed to enrol 12,200 caregiver-infant pairs selected using consecutive or random sampling methods — depending on size of facility. Caregivers agreeing to participate in the study were interviewed, and their infant’s dried blood spot (DBS) was sent in for HIV testing.

If the mother identified herself as being HIV-positive, her infant’s specimen was sent directly for HIV DNA PCR testing. If she did not, her infant’s specimen was first screened with an HIV ELISA test to identify HIV-exposed infants, and PCR was performed on all antibody-positive specimens. (HIV antibody testing will detect all HIV-exposed children because infants inherit maternal antibodies, which may remain in circulation for up to 18 months after birth. HIV DNA PCR can distinguish between infection with HIV and maternal antibodies by detecting HIV DNA.)

Between June and December 2010, 92% of caregivers of eligible infants agreed to participate in the survey, and interviews and DBS results were available for 9915 matched caregiver-infant pairs. This fell somewhat short of the target sample size that the researchers had calculated as necessary to obtain valid national and provincial level transmission. To adjust for the shortfall, the analysis was weighted based upon the sample size achieved in relation to the live births in the province over the study period — though the researchers noted that this assumed that the available sample was representative of the whole province, something less than certain in the Eastern and Northern Cape, which reached less than 60% of desired sample size.

Results

The study found that infants in South Africa are at a very high risk of HIV exposure, but the PPTCT programme in most provinces has become more effective than in the earlier reports. According to the weighted analysis, 31.4% (95% confidence interval (CI) 30.1-32.6%) of all infants born in South Africa are HIV-exposed, ranging between 15.6% (95% CI 13.0-18.3%) in the Northern Cape (though this province was significantly undersampled), and 20.6% (95% CI 16.8-24.9%) in the Western Cape to a high of 43.9% (95% CI 39.7-48.0%) in KwaZulu-Natal.

HIV infection at 4-8 weeks was diagnosed in 3.5% (95% CI 2.9-4.4%) of the HIV-exposed infants nationally, but programme performance varied substantially by province. If the specimen in the Northern Cape was indeed representative, their programme has a low transmission rate of 1.9% (95% CI 0.1-4.5%), while Gauteng also does well with 2.3% (95% CI 1.3-3.3%) — however the Free State and Mpumulanga did substantially worse with 5.7% (95% CI 3.5-7.9%) and 6.2% (95% CI 4.5-7.9%) of the exposed infants testing positive, respectively.

Despite its very high maternal HIV prevalence KwaZulu-Natal has cause to be proud with only 2.8% (95% CI 1.7-4.0%) of its many HIV-exposed infants testing positive, somewhat better than the national average — and worlds away from the 20-21% transmission rate described in earlier reports.

Overall, the study found the weighted national HIV prevalence at 4-8 weeks was 1% (95% CI 0.8-1.2%).

During the survey interviews, the caregivers were asked about whether they had been offered or received the key interventions along the PMTCT cascade — such as HIV testing, the return of their results, if HIV positive whether they received a CD4 cell test, were put on ART, etc.). These data showed variabilities in service coverage and allowed the researchers to analyse the performance and uptake of the various individual interventions in each provincial programe.

Notably, ELISA testing revealed that 4.1% of infants whose mothers reported being HIV-negative were actually exposed to HIV — possibly because the woman had become infected in the time since her antenatal test. This underscores the need to provide repeat testing services in pregnancy and to offer couples testing to encourage the woman’s partner to get tested.

What about late postnatal HIV transmission?

In other studies up to 40% of vertical transmission has been reported to occur in the late perinatal period as a result of exposure to HIV in breastmilk. Transmission primarily occurs when mixed feeding is practised (as opposed to exclusive breastfeeding which poses much less risk). Unfortunately, 18% of the caregivers of HIV-exposed infants reported practising mixed feeding of their HIV-exposed infant in the last eight days before their interview. Again, this varied substantially by region, with the lowest rates in the Western Cape but very high rates in Mpumulanga and Limpopo.

To the extent that they are implemented, South Africa’s current guidelines for the prevention of vertical transmisison should reduce the risk during the late postnatal phase when compared to previous years. Nevertheless, more infants can be expected to become positive and the rate of final HIV transmission will be higher than that reported at 6 weeks.

“This emphasises the importance of tracking MTCT rates until 18 months to measure the effectiveness of the complete PMTCT programme,” said Dr Goga. So in 2011 and 2012, the SAPMTCTE will follow up the infants to measure HIV transmission rates between six weeks and 18 months. The survey will also be repeated each year to track vertical transmission rates over three years, and will ultimately be integrated into the Department of Health’s routine M&E systems.

 “The quality of the PMTCT programme varied across province,” said Dr Goga. Consequently, these results will be shared with each province to show them where there are gaps in their service delivery — such as a failure to provide CD4 cell testing and deliver the results in a timely fashion, failure to get those women who qualify onto ART, or failure to provide adequate counselling or support for safer feeding options. The results will also be used to set new targets for each province

“The SAPMTCTE data shows that virtual elimination of paediatric HIV could be possible by 2015, with intensified effort. However, we need to address the postnatal component and variabilities in PMTCT service coverage,” Dr Goga concluded.

Suggestions of further reductions in HIV transmission at 6 weeks AND 18 months at Kheth’Impilo supported sites

On the day before Dr Goga announced the survey results, a poster presentation from Kheth’Impilo, an NGO that supports health services and community-based adherence interventions at primary healthcare clinics and district hospitals in KZN, Mpumulanga and the Eastern Cape, reported similar findings at supported sites over a period beginning in October 2009 through the end of March 2011, suggesting continued improvement in a number of PPTCT indicators, including reductions in HIV transmission at six weeks and 18 months.5

Kheth’Impilo provides a number of support services to these facilities, including infrastructure support, patient flow adjustment, register development and supply, as well as critical staff including doctors, nurses, pharmacists, social workers and more recently PMTCT quality mentors (QMs), who rotate between several sites in a sub-district. The PMTCT quality mentors coach antenatal clinic staff using quality improvement approaches to identify service delivery gaps, help them identify and try out potential solutions, and assess the outcomes before scaling up implementation (plan, do, study, act). KI also employs a community services cluster — a team of patient advocates and community educators who support adherence to ART and the PPTCT protocol, make certain that patients aren’t lost to follow-up and achieve successful referrals between services. They also encourage testing of partners and household members — including infant testing at six weeks and 18 months.

The poster described programme data aggregated from DHIS reports by quarter, drawn from the monthly reports supplied by each participating clinic. The data were cross-sectional, and are not individual level cohort data, so the denominators used are sometimes estimates. However, the clinics use the data to inform ongoing quality improvement, and the poster stressed that each site’s personnel review and analyse the data closely to make sure they are an accurate reflection of outcomes, and identify and correct gaps in each monthly report before submitting it. Muddying the picture, however, is the fact that the number of sites being supported increased from 14 to 58 over the study period.

That being said, some trends worth noting emerge. There is a slight progressive increase in the women who are already aware of their status when first visiting the antenatal clinic — which the poster authors suggest is possibly a result of the government’s HIV counselling and testing campaign. The sites succeed in getting almost all of the women not already known to be HIV-positive to test at booking.

Moreover, an increasing proportion of women agree to be tested again around week 32 of gestation — from 15.7% in the first quarter to 27.9% in the most recent quarter. This measure is important because it seeks to detect HIV infections that occur during pregnancy.

Coverage of ARVs for PPTCT to women not on ART is almost universal. The percentage of women on ART for their own health has recently increased, largely, the authors say, due to the increased number of women qualifying for ART after the 2010 ART guidelines. Unfortunately, the poster did not report on the percentage of positive women getting their CD4 cell counts before delivery or the percentage of women qualifying for ART who received it — or how long before labour they received it.

About two-thirds of the women agreed to have their infants tested for HIV by PCR at six weeks — and there has been a marked reduction in early transmission since the first quarter of 2009, when 9.7% of the HIV-exposed infants tested positive, to the most recent quarter when 2.4% tested positive.

At 18 months follow-up, the poster reports a “massive undertesting” of infants. Nevertheless, the data here also suggested a substantial decline in HIV prevalence, with 10.7% of 18-month-old infants testing antibody positive in the first quarter, and 3.8% testing positive in the quarter ending in March 2011. While these figures are not a reliable measure of infection in the cohort, given the missing data, the trend over time is encouraging.

The authors note that the difficulty in obtaining timely CD4 and PCR results is a challenge to improving outcomes — and that point-of-care tests are needed that can be performed by a staff member on site who can also counsel patients. The authors state that another challenge for the PPTCT programme is the lack of adequate data management tools to monitor the key indicators and interventions of the programme. Meanwhile staff burnout and the difficulty of providing infant follow-up out to 18 months are perennial problems.

Strategies for improving PPTCT: examples of successful programme improvement

In addition to the quality improvement, mentorship and community-based interventions described above, several other posters provided valuable insights and strategies to improve PPTCT programme performance and outcomes in infants.

Proper data recording itself appeared to be associated with better programme performance in a study in a sub-district in the Northwest Province that evaluated recorded indicators that were associated with proper delivery of infant prophylaxis — which was only successfully provided to 78% of HIV-exposed infants in the sub-district. Significant predictors of successful infant prophylaxis delivery included making certain there was an ANC card in the delivery file and that the mother’s HIV status was properly reported in the Maternity register.6 In addition, documentation that a CD4 cell count had been taken and that PPTCT ARVs were provided to the mother were also significant predictors of whether the infant would be appropriately treated. The authors noted that “each step of the PPTCT cascade properly completed seemed to build momentum towards successful provision of the intervention to the infant.”

The need for better documentation — in particular, patient-held cards clearly documenting maternal and infant HIV status, was also noted in a feasibility study for SAPMTCTE that looked at whether early infant diagnosis services could be provided by primary health clinics, integrated into routine immunisation and child healthcare services, and what sort of linkages existed between early infant diagnosis (EID) and ongoing treatment, care and support services.7

In answer to the latter question, significant problems were identified making successful linkages between the immunisation, PPTCT and maternal child health services. The authors concluded this was because of the lack of a consistent system clearly explaining her and her infant’s situation to healthcare workers at the referral site. The recently released new Road to Health card, which records immunisation and growth to the age of five years had not been taken up widely at the time of the study, but could address some of these issues.

Referral systems for children testing HIV-positive to ART and community-based support services also need to be made more effective. Only 57% of clinics reported having a mechanism to ensure these children made it to ART services, and the ART services also had poor mechanisms for follow-up.

While most of the clinics had trained personnel, in particular professional nurses, who could draw blood for early infant diagnosis, clinic managers identified a clear need for task shifting to trained nurse assistants and lay counsellors. Maintaining adequate stock of dried blood spot kits was an issue for some provinces. The study also noted a need to develop communication strategies to update managers and healthcare providers when programme policies or guidelines are updated.  Notably, a study of participants and healthcare workers in a PPTCT programme in Soweto reached essentially the same conclusion after finding that many of the women, and even a number of their PPTCT nurses, think they have a good understanding of the guidelines, but actually have poor or inaccurate knowledge in sometimes critical areas.8 The authors recommended improving community awareness of PPTCT and mentorship of staff.

One changing policy includes the recent changes in South African regulations to allow nurses to prescribe antiretrovirals. Another oral study presented at the conference decribed how, despite training, nurses at primary healthcare clinics were reluctant to prescribe — and this was delaying access to ART in pregnant women qualifying for treatment.  A separate article in this HATIP on task shifting will describe how quality improvement measures helped nurses in one rural sub-district overcome their reluctance and begin putting pregnant women on ART.

The need to increase community awareness and demand for PPTCT services was also identified in a report describing the experience in sub-districts of the Eastern Cape supported by the international NGO PATH to reduce vertical transmission.9 In addition to improving the quality and access of HIV testing and counselling services during antenatal care, better patient education and support including improved counselling on family planning services and on safer infant feeding practices was also a clear objective. Enhanced community engagement and leadership in the programme was pursued as a strategy to reduce stigma and increase support for HIV-positive pregnant women. Consequently, regular radio talk shows were organised to discuss topics such as pregnancy planning for HIV-positive couples, care of HIV-positive pregnant women and safer infant feeding practices.

At the same time, PATH sought to strengthen the health system (with better data reporting, supply logistics, referral linkages, lab test performance and turn-around and developing training curricula etc). Training and support were also enhanced. PPTCT training was scaled up to all previously untrained staff members and updates provided to trained staff on guideline revisions. This was accompanied by increased monitoring and supervision, as well as clinical mentoring.

PATH employed a quality improvement process at the supported sites. Once the clinic staff had been trained on the guidelines, a meeting was held where service delivery gaps were identified by comparing actual performance to the programme targets. Staff were asked open ended questions to identify the root causes of the gap and potential solutions. These were applied, the results were monitored and analysed on a monthly basis, and implementation went forward depending upon the results.

PATH also got a local cell phone service provider to supply clinics with free SMS bundles, which were used to remind mothers to bring their infant in for early infant diagnosis at six weeks.

The PATH poster provided a brief snapshot of some of the programme results. Antenatal testing rates in the four supported districts is above 100% (due to repeat testing). The HIV-positive test rate in HIV-exposed infants ranges from 3.1% to 4.7%, but on average appears to have steadily declined from close to 8% in September 2008 to less than 4% for the period concluding September 2010.

While not as rigorous or extensive as the SAPMTCTE survey, the results again appear to be consistent.

Pegging the increasing effectiveness of the PPTCT programme in South Africa to any single intervention may not be possible or sound at this point. Indeed, it is probably premature to attribute it to raising the threshhold for pregnant women to start ART to 350 CD4 cells. Although this should ultimately lead to even lower transmission rates, it is not clear that this policy was in place long enough to have led to a profound change in the actual percentage of women getting on ART in time to make a difference — while they are still pregnant. Uptake of ART is still delayed by challenges providing access to CD4 cell tests at many primary healthcare sites, delays in getting the results, and the reluctance of nurses at some primary healthcare clinics to prescribe treatment.

So how did they do it? Perhaps more credit should be given to the many healthcare providers involved in South Africa’s PPTCT programme, who have come up with and implemented these various interventions, which alone may only improve performance incrementally, but together amount to a significant number of infections averted, and a triumph for South Africa.

"I think it's sending out quite a positive message because I think there has been a lot of disappointment and morale has been quite low in terms of the effectiveness of programmes,” Dr Goga said in an interview with Australian Broadcasting Company. “And I think here we have an example of success.”

20,000+ Partnership contributes to dramatic reduction in PPTCT in KwaZulu Natal

There were a couple of oral presentations to suggest that it was the efforts of health workers to improve the quality of the PPTCT service they were providing, that may have been largely responsible for the reductions in vertical transmission being seen in the country.

“Quality improvement approaches have helped to close the service delivery gaps for PMTCT interventions and seem to be contributing to the reduced number of HIV infections in children in KZN,” said Dr Wendy Dhlomo-Mphatswe, of the 20,000+ Partnership.10

The claim appears to be supported by evidence showing concurrent improvement in a number of indicators of performance at different steps of the PPTCT cascade.

HIV accounts for 50-60% of hospital deaths in children and 30-40% of all childhood deaths in KwaZulu Natal, with poor health system performance negatively affecting perinatal transmission rates. As already noted, previous studies suggest the rates of HIV diagnosis at first immunisation was above 21%.

The KZN Department of Health and the University of KwaZulu Natal developed the 20,000+ partnership. The Institute of Healthcare Improvement (IHI) which specialises in health systems intervention, and using data for quality improvement, contributed experienced staff to the project team.

The partnership aims to decrease the vertical transmission of HIV to infants a target of below 5%; this is the target set by the South African National Strategic Plan 2007-2011, and to improve overall child survival in KZN through health systems support interventions using Quality Improvement methods. The partnership is working across the three districts of Ethekwini, Umgungundlovu and Ugu, which each have an antenatal prevalence above 40%. 20,000 is the number of HIV infections in infants that can be prevented each year in KZN if every pregnant HIV-infected woman receives care according to the National PMTCT guidelines.

The partnership was tasked to work with the existing staff to improve the health system performance in the PPTCT programme, and to use routine health information (DHIS).  The project involves 15 hospitals, over 200 clinics serving a population of around 5 million people, and runs from April 2008 until April 2013. In addition to the reducing PPTCT, the partnership’s objective was to develop health systems improvement capacity at provincial, district and facility level that could sustain effective PMTCT programmes, but that could be applied to any other health domain and service delivery.

The partnership staff worked with improvement teams and leadership at every level. These included the district information office data team that fed data back and forth with the clinic and hospitals team, and each of these teams was in turn supervised by the district task team. The district task team reported back to the district leadership which reported to the provincial leadership.

The quality improvement process

“Quality Improvement” in healthcare is a simple method to identify gaps in the healthcare system, and a systematic way to close those gaps and safely improving the process of care,” said Dr Dhlomo-Mphatswe. She said the project sought to “apply local wisdom; focus on the data (stop the blame game); work `smarter` not just `harder`, and that partnership and teamwork is the only way forward.”

The process begins with discussing the problem with the team, performing a system analysis of the activity or process to be affected. The team then exchange ideas about possible solutions, and choose a good one to move forward. Then the team follows the cycle of ‘plan, do, study, act.’ A plan is drawn up of what will be tried, how will it be done and by whom. Then the team ‘do’ the intervention, and ‘study’ the results closely. If the idea results in improvement, then the ‘act’ upon it. Then implementation goes forward to see whether the intervention continues to succeed and is sustainable.

In the case of PPTCT, they first needed to map out the processes of PMTCT care and track progress with data. For PPTCT, these are many, starting with the first ANC visit, then HIV counselling and testing, HIV positive women are then supposed to have get a CD4 cell test, the results of which determine whether she is referred to ART, or provided with AZT, while the infant is given nevirapine prophylaxis, and six weeks after childbirth (around immunization), the caregiver and child pair should return for early infant diagnosis.

There should be data to track each part of this process resulting from data recording and reporting from each site. Then the data are processed and sent back in a form that allows the clinic team to see how they are doing, which in turn makes them realise the value of their data (and perhaps improve their data recording practices). Initially however, there were problems with the data, so the partnership then first had to focus on data improvement initiatives. Once that issue was take care of, the teams could compare their performance with targets.

Dr Dhlomo-Mphatswe displayed graphs showing improvement in a number of indicators. For instance, the goal is to get all the women who come into the antenatal clinic tested for HIV at their first visit but when the project began there was a significant gap. Over time, that gap has been dramatically reduced. Likewise, there were significant increases in the number of women referred for ART, as all the districts can online, a subsequent increase during the ART campaign, and a further increase when ART started being initiated at the primary health centre.

The initiation of life-long ART for eligible women is following the same trend upwards — with a increase as the new art guidelines were released, though not, at first, when ART first started being initiated at the primary health centre. It turned out that despite training and now being qualified to prescribe, nurses still lacked confidence to do so. A separate presentation discussed in more detail in next week’s HATIP described quality improvement of nurse-initiation management of ART in the Ugu district.11 From June –September 2010, the percentage of clinics initiating eligible pregnant women on ART increased from 32% to 100% by December 2010.  The absolute numbers also increased dramatically. Overall, though, the districts being supported by the 20,000+ Partnership are still about 25% shy of their targets for starting eligible pregnant on ART.

Early infant diagnosis remained mostly stable during the first year of the project, however, there are clear improvements when two of the districts began to focus on PCR from just under 40% to about 70%, and another jump when the infant testing guideline changed, to reach about 80% of exposed infants.

At roughly the same time as improvements of these indicators, the 6-week HIV transmission rate in Ugu district was declining, from 12% in 2008, to 8% in 2009, to the target of 5% in the first quarter of 2010 (and with about 90% of the HIV-exposed infants being tested for HIV).

Implications of a quality improvement approach on the health system

While the PPTCT regimens also improved significantly over this period, it would have had little impact if the poor performance of the health service persisted. For instance, changing the CD4 cell criteria on which to initiated ART or allowing the nurses to initiate ART doesn’t matter if women are being referred for ART — which was the situation previously, or if nurses don’t prescribe. Rather these quality improvement measures were synergistic and likely essential for outcomes to improve to the great extent they did.

“The quality improvement approach is a multi-faceted and multi-disciplinary approach that effectively improves the working and coordination of the health system through engaging the leaders, and the workforce to be part of the solution,” said Dr Dhlomo-Mphatswe. “Data is used to guide improvement — and the end-product is better use of existing resources…. And in addition to reducing the number of HIV infections in children in KZN, the same approaches can be applied to other areas of MCH care and help strengthen health systems.”

Mothers2mothers: mothers helping mothers to save babies

by Lesley Odendal

An innovative method used in sub-Saharan Africa to support HIV-infected mothers to prevent the vertical transmission of HIV was presented at the 29th International Confederation of Midwives Congress held in Durban, South Africa, from June 20-23 2011.

Being supported by HIV-infected women who had been through the prevention-of-mother-to-child (PMTCT) programme themselves was found to yield better PMTCT uptake and psychological outcomes than when HIV-infected women had not been supported through the programme.

mothers2mothers (m2m) is a community-based organisation made up mothers living with HIV who have personally received PMTCT care. These women, known as mentor mothers, provide education and support for pregnant women and new mothers living with HIV. By working side-by-side with doctors and nurses in public health care facilities, mentor mother assume responsibility for ensuring that patients understand, accept, and adhere to interventions that are prescribed in the PMTCT programme.

A 2007 study which evaluated the results of the programme in KwaZulu-Natal yielded good results. Women who had been exposed to the m2m programme, reported significantly higher rates of having undergone CD4 testing during their last pregnancy (79 percent vs. 57 percent in non-participants; p < 0.01) and knowing their CD4 count after testing (88 percent vs. 72 percent; p < 0.01) than non-participants.

95 percent of all mothers in the m2m programme received nevirapine during labour (the standard of care in the PMTCT programme at the time) and 88% of infants received nevirapine. 89% of all women who had been involved in the programme we found to practice exclusive feeding methods of the infant to prevent vertical transmission of HIV. 70 percent of the women were also found to be using contraception after the birth of their baby.

Among postpartum women, women who had been through the m2m programme  were significantly more likely to have disclosed their HIV status to someone than non-participants (97 percent vs. 85 percent; p < 0.01), and to have done so prior to delivery.

The programme evaluation by the Population Council used a pre-post, quasi-experimental study design to assess program effectiveness. There were three evaluation sites in the Pietermaritzburg area of KwaZulu-Natal where women from urban and peri-urban settings were recruited from. The eligibility criteria for the study included being between the ages of 18 and 49 years, knowing one’s HIV status, and either 6–9 months pregnant or 12 weeks or less postpartum.

Two cross-sectional surveys were conducted. The first survey, conducted in 2005, collected baseline data before  m2mwas introduced. 183 HIV-positive pregnant women and 178 HIV-positive postpartum women were interviewed using a structured questionnaire.

The second survey was conducted in 2006, one year after m2m was introduced. 345 HIV-positive pregnant women and 350 HIV-positive postpartum women were interviewed using the same questionnaire but with additional questions about program exposure and interaction.

The study also demonstrated psychosocial benefits of the m2m programme. Pregnant women from the m2m programme were significantly more likely to report feeling they could do things to help themselves, cope with taking care of the baby, and live positively in comparison to non-participants.

Postpartum participants reported feeling less alone in the world, overwhelmed by problems, and hopeless about the future compared to women who had not been exposed to the m2m programme.

There are 348 public health facility sites in South Africa where 890 m2m staff are working. The programme has also spread to eight other countries in sub-Saharan Africa where mentor mothers are working in 366 sites, including in Kenya, Lesotho, Malawi, Rwanda, Swaziland, Tanzania, Uganda and Zambia.

m2m is based on the concept that “peer support is an optimal model for effective education and social empowerment, and that mothers themselves are the best vehicle to provide support to other mothers.” Mentor mothers provide one-on-one counselling for pregnant women, and also facilitates support groups in the health facilities. Mentor mothers also make daily visits to the labour and delivery wards to speak with expecting mothers who may be inpatients or newly delivered mothers who are awaiting discharge.

Another benefit of the programme is that by becoming a mentor mother, employment and professional development is given to local women living with HIV.

Each mentor mother participates in two weeks of training that covers basic medical knowledge about HIV infection and antiretroviral therapy, behaviours that help prevent mother-to-child transmission, safer feeding options for infants, counselling methods that can help women to disclose their status, strategies for negotiating safer sexual practices, and nutritional guidelines for women living with HIV.

Reference

Baek C et al. Key findings from an evaluation of the mothers2mothers program in KwaZulu-Natal, South Africa. Presented to: 29th International Confederation of Midwives Conference, Durban, 2011.

References

[1] Goga A, Dinh T-H, Jackson D. Results of an evaluation of effectiveness of the National PMTCT Programme at six weeks postpartum, SA. 5th South African AIDS Conference, Durban, 2011.

[2] Jackson DJ, Chopra M, Doherty TM et al. Operational effectiveness and 36 week HIV-free survival in the South African programme to prevent mother-to-child transmission of HIV-1. AIDS 21(4):509-16, 2007.

[3] Rollins N, Little K, Mzolo S et al. Surveillance of mother-to-child transmission prevention programmes at immunization clinics: the case for universal screening. AIDS. 21(10):1341-7, 2007.

[4] Rollins N, Mzolo S, Moodley T et al. Universal HIV testing of infants at immunization clinics: an acceptable and feasible approach for early infant diagnosis in high HIV prevalence settings. AIDS. 23(14):1851-7, 2009.

[5] Grimwood A et al. Encouraging PMTCT outcomes over an 18 month period, Q1 October 2009 to the end Q2 March 2011 at Kheth’Impilo (KI) supported PHC and district hospitals. 5th South African AIDS Conference Durban, 2001.

[6] Matson CA, Kraai GB, Moultrie H. Coverage and predictors of ARV prophylaxis to HIV-exposed infants at the time of delivery in a rural sub-district of South Africa. 5th South African AIDS Conference Durban, 2001,

[7] Woldesenbet S et al. Capacity of primary healthcare (PHC facilities to integrate EID into routine child health care: results of a situational assessment conducted prior to the South Africa PMTCT Effectiveness Survey (SAPMTCTE), 2010. 5th South African AIDS Conference Durban, 2001.

[8] Mnyani CN et al. Translating theory into practice — lessons learnt from the Soweto, South African, PMTCT programme. 5th South African AIDS Conference Durban, 2001.

[9] Dana N. Strategies to reduce mother to child transmission of HIV and AIDS in selected sub districts of South Africa’s Eastern Cape Province. 5th South African AIDS Conference Durban, 2001.

[10] Dhlomo-Mphatswe. Using quality improvement to reach public health targets. 5th South African AIDS Conference, Durban, 2011.

[11] Shibe M. Quality improvement can empower nurses at PHC facilities to deliver ART.5th South African AIDS Conference, Durban, 2011