In the literature, empirical work on PMTCT in India (N

=

134) seems to be dominated by epidemiological studies (N

=

60). The 46 papers related to the utilization/provision of the cascade of PMTCT services were mostly from the four high prevalence states in southern India and from the public sector. Of these 46 papers, 20 were related to experiences of implementing a PMTCT program and 26 were related to individual components of the cascade of PMTCT services.
Studies related to experiences of the implementation of PMTCT programs have been largely restricted to providing a numeric summary of the program with some efforts to understand the cost effectiveness of the programs and barriers in providing PMTCT services particularly counseling and HIV testing. The current literature shows high rates of drop out especially among HIV infected women before they could receive ARVs. Studies also suggest that the cost effectiveness of PMTCT programs can be improved if they are integrated with other services and if PMTCT services can demonstrate other health benefits than just preventing HIV transmission. Lack of knowledge about the availability of HIV testing facilities among women and fear of stigmatization from health providers seem the most important barriers in accessing counseling and HIV testing services. High rate of testing in private health facilities - where PMTCT program are not implemented - also suggest the possibility of uninformed testing of pregnant women in these facilities. However, there is a lack of literature on coverage of PMTCT services in private health system. Research is also needed on systematic program evaluation documenting potential health benefits of ANC counseling to the mother and the child and assessing the outcome of PMTCT program by measuring outcome indicators such as HIV free survival among children and linking HIV infected women and children to care. Literature on best practices in the field setting that are shown to improve uptake of PMTCT services is lacking as well.
Our review on the individual components of the PMTCT cascade showed that there is a lack of literature on factors associated with pregnant women’s access to ANC care, HIV infected women’s immunological assessment (using CD4 testing) for evaluating their eligibility to receive combination ART or HAART, feasibility and cost effectiveness of repeat HIV testing among pregnant women who are in third trimester of their pregnancy, feasibility and uptake of early infant diagnosis and factors related to linking HIV infected women and children to postnatal care. Studies on HIV counseling and testing of pregnant women suggest that it is acceptable to women and feasible to provide in both public and private health facilities. The studies on infant feeding show high morbidity among non breast-fed infants. However, high proportion of HIV infected women in these studies had opted for replacement feeding (mainly animal milk). Among women who chose to breast-feed, practicing exclusive breastfeeding was observed to be challenging.
Except for the papers providing numeric summary of the uptake of PMTCT services, the overall quality of the papers was good. However, generalizability of the findings particularly for studies related to cost effectiveness that are restricted to a single state and studies on infant feeding that are carried out in a single tertiary care hospital might be low. The studies on the numeric summary of uptake of services did not provide data on all the components of the cascade of PMTCT services and the sample sizes for some studies were low [
19,
27,
34], which made it impossible to summarize the percentage drop-outs at each step of the cascade.
The rates of uptake of PMTCT services and the findings of cost effectiveness of programs varied across studies. While the variations in the cost across studies on cost effectiveness could be because of the comparison of different programs and services in these studies, the variations in rates of uptake could be because of the different program strategies or due to lack of consistency in the methodology for estimating the uptake rates. For example, in the study by Joshi et al. (2010) the reported ARV uptake was 92%. However, this high rate was mainly because the data of only the women who delivered in the hospital were considered for estimating the ARV uptake among mother baby pair leaving out 36% of the HIV infected women who were lost to follow up. The high rate of uptake in the two studies by Parameshwari et al. (2009) and Panditrao et al. (2011) (reported under barriers in accessing care) - both part of the Elizabeth Glaser Pediatric AIDS Foundations’ PMTCT program in India - might be because of the program implementation strategies, such as quality of counseling, educational material used, monitoring and evaluation systems and demographic and sociocultural factors [
20] of women enrolled in these programs. Similar to most papers describing programmatic experiences the paper by Parameshwari et al. (2009) was based on a descriptive analysis of a small number of HIV infected women (N

=

56) whereas Panditrao et al. (2011) analyzed program data from 2002 to 2008 of 734 HIV infected women who were enrolled during pregnancy and 770 HIV infected women who reported a live birth in order to understand factors associated with loss to follow-up (LTF) before delivery and after delivery respectively using multivariate statistical techniques. Most of the current literature on program experiences does not describe the program implementation strategies which limit analysis on their association with the uptake. There is an urgent need for a systematic understanding of these factors in India and to improve the reporting of existing program data to identify strategies and practices that could improve efficacy of PMTCT program.
In this review, data about the program from public as well as private health facilities suggest that a substantial proportion of HIV infected women enrolled in a PMTCT project drop out, without receiving the complete cascade of services. Loss to follow-up of women enrolled in PMTCT programs has been recognized as a major factor affecting the efficacy of the program in resource-poor countries [
64,
65]. While there have been some studies from Africa on risk factors relating to loss to follow-up, a systematic understanding of the factors affecting continued access of services among HIV-infected Indian women is lacking [
20].
Although this review on PMTCT literature in India has uncovered only limited evidence regarding factors affecting the uptake of counseling and HIV testing among pregnant women - such as the availability of a HIV testing facility, discussion about HIV testing with care providers and perceived stigma from the health facilities -, elsewhere there is burgeoning literature on the factors affecting such uptake as well as the integration of PMTCT programs with other health services. A systematic review of factors affecting counseling and HIV testing, mainly from Africa, suggests that provider-initiated counseling and testing, group pre-test counseling sessions and rapid HIV testing with same-day results increase the uptake of counseling and testing [
66]. However, another systematic review of South African studies suggests that in addition to these factors quality of counseling and contextual factors such as HIV prevalence, human and physical resources significantly affect counseling, and the review recommends that best available evidence from research studies regarding these factors should be used to develop guidelines for the local context [
67]. While a family-centered approach in implementing PMTCT and integration of the PMTCT program with other services has been generally recommended in many countries including India [
17,
68], a recent review by Cochrane on the efficacy of integrating PMTCT with other services argued that evidence from resource-poor countries does not allow one to conclude that integrated care is more effective than non-integrated or partially integrated care. The influence of the socio-cultural context in the uptake of services highlighted in these studies strongly suggests the need to conduct research locally to determine the factors affecting the efficiency of PMTCT services and the benefits of integrating PMTCT with other services [
69].
This review has shown that evidence on experiences in providing PMTCT services in the private health care system is limited as compared to public health care system. This is striking as, in India, the private health care system plays a significant role in the delivery of health care. Approximately 70 percent of urban households and 63 percent of rural households access care from the private health sector [
70]. Though the role of the private sector is recognized in the third National AIDS Control Program (NACP-3) [
71], efforts to include it have been sporadic. There is a complete lack of systematic knowledge on the current coverage of PMTCT in the private sector and on the referral mechanisms for women detected as HIV-infected in the private sector. There is an urgent need for assessing the most effective models for public-private partnerships (PPPs) in the country.
The published public health literature also seems to be geographically skewed and dominated by the analysis of data from a few centers. Of the 42 studies using data at the district/state level, 16 papers were from Maharashtra (of these, eight were from a single tertiary care center in Pune), eight from Tamil Nadu, seven from Andhra Pradesh (out of which five were cost effectiveness studies) and four were from Karnataka. It is important to note that in spite of being one of the high prevalence states since the beginning of AIDS epidemic in India, nothing has been published from Manipur and Nagaland. The composition and performance of health systems in these two north-eastern states differ from the other four high HIV prevalence states in the country. As the provision and utilization of PMTCT services depends on sociocultural and health-system-related factors, understanding of these factors across states would be essential for effective implementation of the program in the entire country. It is also interesting to note that except one all the papers on cost effectiveness are authored by Dandona et al. Cost effectiveness analysis is one of the important tools for program planning and implementation. Therefore allowing a richer academic discourse on this issue by attracting and training more scholars in this field might be beneficial.