In this study there were a series of missed opportunities that led to women not receiving nevirapine according to the national PMTCT protocol[4
]. A quarter (15) of the women reported not taking nevirapine. Six of the women did not get tested and were of unknown status prior to delivery. Reasons for not testing were mainly health systems failures such as non-availability of counselors and supplies. Two of the 15 women did not receive their HIV test results until after delivery. In both instances the women had attended antenatal care on numerous occasions yet health workers failed to give them NVP in time. Seven of the 15 women tested, received results but did not take NVP not because of stigma, ignorance but because of the immediate context of the birth process in addition to health system failures.
Fear of knowing one's HIV status and disbelief of test results have been described previously as important reasons for drop out from PMTCT services [13
] however, in our sample this explained only a minority of missed opportunities. Evidence from other programmes suggests that such fears are likely to be reduced further as the programme becomes more established[15
]. Instead it was health system constraints related to testing and the provision of results that were the key reasons for missed opportunities. A lack of counselors and testing equipment was found to be prevalent across the country during early evaluations of the PMTCT pilot programme [12
Of great importance is the functioning of the health system. 40% of our respondents had not been tested due to health systems failures. HIV testing serves as an entry point to PMTCT. A weak health system allows for leakage in essential steps such as HIV testing, which undermines the intervention. Strengthening testing uptake and logistics is urgently required. However, it is important to note that these missed opportunities occurred within the context of opt-in VCT. Routine offer (opt out) of HIV testing within antenatal care (i.e. antenatal HIV testing is part of routine screening for infections, including hepatitis B, syphilis and rubella) has been advocated. There is evidence from various studies which demonstrates that an opt-out approach to VCT identifies a greater proportion of those infected [16
] and provides greater opportunities for HIV care and treatment.
However, the discussions on the VCT opt-out approach assume that the quality of care in all settings is optimal. Findings from developing countries do not support this assumption. For instance a recent publication from South Africa showed only 28.6% of women from a rural site had a syphilis test performed [5
]. Given these results it is unclear how the VCT opt-out approach would address the current missed opportunities occurring in the VCT opt in approach.
The second reason for missed opportunities in this study was health system constraints related to mothers not receiving their results. The testing kit used in this context is the rapid HIV test. It is therefore unclear why women were not given their results and this issue requires further exploration.
The last groups of women were those who knew their status and also knew they had to take NVP but still did not take it. For reasons related to tablet provision and instruction giving, in many cases the problem stemmed from poor communication and a lack of a locus of responsibility. Improving communication could reduce these missed opportunities. On the otherhand the locus of responsibility presents a challenge. According to the PMTCT protocol women should get NVP during antenatal care to self administer at the onset of labour. However, in this group the responsibility of administering the NVP fell more with the healthcare workers. Furthermore 92% of deliveries in South Africa are attended by trained health personnel. But even in this context women failed to receive NVP [21
The apparent simplicity of the present NVP regimen gives women only one opportunity to reduce transmission and this opportunity is too often missed. The first-line regimen suggested by WHO is either a triple or dual combination short course regimen from 32–36 weeks of pregnancy through labour and delivery and for one week postpartum to mother and infant [22
]. We argue that the recently revised WHO recommendations for a more efficacious short course regimen may, despite its apparent complexity, actually reduce missed opportunities, among our study women 7 out of the 15 missed opportunities could have been averted with a multi dose regimen. If one or two doses of this short course are missed, the implications are less serious in terms of efficacy than if the single NVP dose is missed, though missing a dose in the short course may have implications for future drug resistance[22
Evidence from studies of other diseases suggest that the more complex regimen may also result in improved adherence as studies show higher adherence with multiple doses such as daily doses instead of erratic doses such as once or twice weekly[23
]. Providing a regimen that starts early in pregnancy should also be feasible as South Africa has an antenatal attendance rate of 90% and a mean number of ANC visits greater than three[10
Whilst the 7 out of 15 women would have benefited from the complex regimen, six out of 15 women who were not tested would not have. It is clear that health system strengthening is essential for the success of any new interventions irrespective of whether the intervention is simple or not. Greater resources, management and integration with routine maternal and child health care have been recommended to reduce these shortcomings and the need for this is further highlighted by the findings of this study.
The limitations of this study were that the data was only collected from the women's point of view. Collecting data on only the women's point of view was an appropriate step given that national evaluations [8
] had demonstrated quantitative missed opportunities. In fact a local evaluation[24
] had cautioned against looking at each step of the programme independently for example paying attention to individual variables such as number of women who received counseling, of those who received counseling the number tested and the number that received the results. This type of analysis masks the cumulative effect of these leakages. This study has shed light on the nature of missed opportunities and health systems leakages on the continuum of care for PMTCT. These could be further explored through collecting data from the health providers. The health provider perspective will be a useful follow-up step in order to ascertain health providers understanding of the underlying reasons for the missed opportunities which arose from health systems failures.
Finally monitoring and evaluation of the quality of care even in the context of dual/triple therapy from different perspectives (Patient and Provider) will be essential in identifying barriers for PMTCT.