The impact of programs for prevention of mother-to-child transmission of HIV on health care services and systems in sub-Saharan Africa - A review

Background The global scale-up of Prevention of mother-to-child transmission (PMTCT) services is credited for a 52% worldwide decline in new HIV infections among children between 2001 and 2012. However, the epidemic continues to challenge maternal and paediatric HIV control efforts in Sub Saharan Africa (SSA), with repercussions on other health services beyond those directly addressing HIV and AIDS. This systematised narrative review describes the effects of PMTCT programs on other health care services and the implications for improving health systems in SSA as reported in the existing articles and scientific literature. The following objectives framed our review: To describe the effects of PMTCT on health care services and systems in SSA and assess whether the PMTCT has strengthened or weakened health systems in SSA To describe the integration of PMTCT and its extent within broader programs and health systems. Methods Articles published in English and French over the period 1st January 2007 (the year of publication of WHO/UNICEF guidelines on global scale-up of the PMTCT) to 31 November 2016 on PMTCT programs in SSA were sought through searches of electronic databases (Medline and Google Scholar). Articles describing the impact (positive and negative effects) of PMTCT on other health care services and those describing its integration in health systems in SSA were eligible for inclusion. We assessed 6223 potential papers, reviewed 225, and included 57. Results The majority of selected articles offered arguments for increased health services utilisation, notably of ante-natal care, and some evidence of beneficial synergies between PMTCT programs and other health services especially maternal health care, STI prevention and early childhood immunisation. Positive and negative impact of PMTCT on other health care services and health systems are suggested in thirty-two studies while twenty-five papers recommend more integration and synergies. However, the empirical evidence of impact of PMTCT integration on broader health systems is scarce. Underlying health system challenges such as weak physical and human resource infrastructure and poor working conditions, as well as social and economic barriers to accessing health services, affect both PMTCT and the health services with which PMTCT interacts. Conclusions PMTCT services increase to some extent the availability, accessibility and utilisation of antenatal care and services beyond HIV care. Vertical PMTCT programs work, when well-funded and well-managed, despite poorly functioning health systems. The beneficial synergies between PMTCT and other services are widely suggested, but there is a lack of large-scale evidence of this.

1. To describe the effects of PMTCT on health care services and systems in SSA and assess whether the PMTCT has strengthened or weakened health systems in SSA 2. To describe the integration of PMTCT and its extent within broader programs and health systems.
Methods: Articles published in English and French over the period 1st January 2007 (the year of publication of WHO/UNICEF guidelines on global scale-up of the PMTCT) to 31 November 2016 on PMTCT programs in SSA were sought through searches of electronic databases (Medline and Google Scholar). Articles describing the impact (positive and negative effects) of PMTCT on other health care services and those describing its integration in health systems in SSA were eligible for inclusion. We assessed 6223 potential papers, reviewed 225, and included 57.
Results: The majority of selected articles offered arguments for increased health services utilisation, notably of ante-natal care, and some evidence of beneficial synergies between PMTCT programs and other health services especially maternal health care, STI prevention and early childhood immunisation. Positive and negative impact of PMTCT on other health care services and health systems are suggested in thirty-two studies while twenty-five papers recommend more integration and synergies. However, the empirical evidence of impact of PMTCT integration on broader health systems is scarce. Underlying health system challenges such as weak physical and human resource infrastructure and poor working conditions, as well as social and economic barriers to accessing health services, affect both PMTCT and the health services with which PMTCT interacts. (

Continued from previous page)
Conclusions: PMTCT services increase to some extent the availability, accessibility and utilisation of antenatal care and services beyond HIV care. Vertical PMTCT programs work, when well-funded and well-managed, despite poorly functioning health systems. The beneficial synergies between PMTCT and other services are widely suggested, but there is a lack of large-scale evidence of this.  [1]. With adequate efforts, more funding and closely monitored progress, the United Nations Programme on HIV and AIDS (UNAIDS) still reported 160,000 new paediatric HIV infections in 2016 [2]. Despite significant progress, the epidemic continues to challenge maternal and paediatric services in Sub Saharan Africa (SSA), with repercussions on other health services beyond those directly addressing HIV and Acquired Immunodeficiency Syndrome (AIDS) [3].
This systematised narrative review describes the impact and integration of PMTCT programs on other health care services. We begin with an overview of prevention of mother-to-child transmission of HIV in the context of global AIDS control efforts, as well as an introduction to the concepts of integration of "vertical" or disease-based programs into broader health systems. We then present and discuss the findings of our review after having presented the methodology.
Overview: HIV, PMTCT, and health system integration Preventingand eliminatingmother-to-child transmission of HIV The highest prevalence of HIV infection is in sub-Saharan Africa, where rates of both prevalent and new infections are consistently higher among women than men and most women are now diagnosed during pregnancy or at delivery through antenatal and perinatal care [4]. The Mother-To-Child Transmission (MTCT) of HIV refers to the spread of HIV from an HIV positive woman to her child either during pregnancy, childbirth (labour, delivery) or breastfeeding. MTCT is the most common mode of transmission of HIV to children. Over 90% of children are infected with HIV through MTCT [5]. The interventions aiming to prevent paediatric HIV/AIDS and for better health of both mothers and their children are known as "prevention of mother to child transmission of HIV" or PMTCT [6].
Since MTCT can be prevented through antiretroviral therapy (ART), a global campaign for its virtual elimination was launched on 21 May 2009 as one of the top priorities for the World Health Organisation (WHO), the Joint United Nations Programme on HIV and AIDS (UNAIDS), the United Nations Children's Fund (UNICEF) and the United Nations Population Fund (UNFPA) [7]. The policy of eliminating new paediatric HIV infections depends on countries reaching not only high rates of initiating ART among pregnant HIV-infected women but also maintaining and supporting them to adhere to treatment [8]. The worldwide expanded access to PMTCT services prevented more than 670,000 children from acquiring HIV from 2009 to 2012 [17]. In 2012, over 900,000 pregnant women living with HIV globally accessed PMTCT services -a coverage of 62% -and only 160,000 new paediatric HIV infections were reported in 2016, compared to 300,000 in 2010 [2]. Four African countries (Botswana, Ghana, Namibia and Zambia) had achieved 90% PMTCT coverage [17] while the PMTCT coverage was over 90% and over 80% in Rwanda and South Africa respectively by 2014 [18,19].
Based on these promising PMTCT figures in SSA, it is possible to envisage achieving virtual elimination of MTCT. In addition, one could expect to see effects of PMTCT programs on other health care services because PMTCT is now largely implemented through existing maternal health programs and services. However, effects on health services and systems are likely to be uneven and complex.
PMTCT is a complex intervention with many involved actors and policies, flows of knowledge, materials, technologies and funds, interacting at global, national and local levels [20,21]. Beyond the operational challenges of actually delivering the cascade on the ground, this complexity and the history of HIV control programs globally raise systemic and political issues related to the involvement of external funders, experts and manufacturerssometimes called the global AIDS industry -in funding and implementing HIV services in low and middle-income countries [22]. In addition, PMTCT programs, especially those which aim to be integrated in and help to strengthen health services and systems, seek to address three health goals, each of which is itself a major effort involving different actors, structures and health strategies. These three goals are: combatting HIV/AIDS, reducing child mortality and improving maternal health [12,23]. Since health services in many countries are organised, funded and managed to deal with distinct diseases and populations but are seen as parts of an overall health system, these multiple interfaces raise questions of whether PMTCT programs have an overall weakening or strengthening impact on national health systems, or no impact at all beyond PMTCT services. For example, Rwanda's success in scaling up paediatric HIV services through effective utilisation of health resources may offer lessons for other developing countries with high prevalence of maternal and paediatric HIV. This successful integration of PMTCT in Rwanda may be attributed to its health system organisation, despite the weakness of the system [24]. Even so, in contrast to these reported good PMTCT outcomes in countries with a history of stagnating health systems like Rwanda (and most other SSA countries), there is little evidence-informed meaningful discussion about possible PMTCT effects on other maternal and child health services and on overall health systems. This review examines some of these questions.

PMTCT integration
If we conceptualise PMTCT as a service within a "continuum of care" approach addressing different periods of women's and children's life course, PMTCT programs could be seen to have potential beneficial synergies with other reproductive, maternal and child health interventions such as Sexually Transmitted Infection (STI) control, early childhood immunisation, antenatal and delivery care, family planning, nutrition supplementation [16]. For example, maternal deaths have been declining at 2.3% worldwide but this reduction is at only 1.7% in SSA where 9% of all maternal deaths were estimated to be due to HIV/AIDS. The slower rate of decline in maternal deaths in SSA has stimulated interest in a range of recent strategies to increase the coverage of good health care, notably free or very low-cost obstetrical care [25]. The improvement in HIV-related care of pregnant women should help to increase the impact of these other strategies, and thus contribute to accelerated decreases in maternal mortality in high-HIV burden countries.
The potential impact of PMTCT on health services may go beyond maternal and child health specialities and may involve other health care services indirectly, with potential opportunities to enhance overall quality of care but also posing threats such as brain drain and redirection of resources from programs not related to HIV in contexts where there is already weak infrastructure. This raises questions regarding the appropriate approaches to address the challenges regarding the accessibility, equity, and quality of healthcare in the efforts to facilitate service delivery and strengthen health systems [26].
The global monitoring framework and strategy put in place towards elimination of new HIV infections in children [12], reflects this wider perspective of integration across programs and services and calls for comprehensiveness through seven priority areas. These priorities are: (i) Ensure leadership and country ownership; (ii) Improve coverage, access and utilisation of services; (iii) Strengthen quality of Maternal, New born and Child Health services to deliver effective PMTCT of HIV and syphilis interventions; (iv) Enhance provision of linked services; (v) Strengthen human resource capacity, supply chain management and information systems; (vi) Improve measurement of performance and impact and (vi) Develop and engage community systems [12]. This framework does not take PMTCT as a self-sufficient entity but instead calls for seeing PMTCT as integrated within a health system. It thus echoes the WHO-recommended integration of PMTCT programs with other healthcare services to achieve the more accessibility and improvement of health interventions [27]. But what does integration entail, analytically and in practice?

Health system integration: Vertical, horizontal and diagonal approaches
The 2003 work of Oliveira-Cruz et al. on the synergies between vertical and horizontal health interventions [28], remains relevant today. They define vertical programs as free-standing programs delivering health care services, put in place to deal with a particular disease or condition with predetermined goals and designed for a known time frame and calculated financial means. Horizontal programs refer to service delivery through the infrastructure of the regular healthcare system [28]. While exploring the impact of vertical programs on health systems and experiences of integrating these programs, these authors encouraged the concerted use of both modes of delivery, according to the capacity of a health system as it changes over time [29]. The combination of both delivery models suggests what Julio Frenk and others refer to as a "diagonal approach" [15]. The "diagonal approach" has been referred to as using a disease-specific intervention (e.g. HIV/PMTCT) to strengthen a general health system [15]. Working together across initiatives can bring more benefits; for example, the major investments in vertical programs addressing specific diseases or conditions such as HIV can through diagonal financing cross-subsidise other programs and also the overall functioning of the health system because "horizontal" services like labs and human resources are essential to implement the AIDS, TB and malaria programs [15].
There is no internationally recognised definition of integrated care and to further complicate analysis, some researchers who address what could be considered as integrated service delivery do not use "integration" as a keyword [27]. Integrated care describes a range of organisational arrangements with variable nature and intensity and comprises two main concepts: a) an organisational structure focused on economic benefits, notably efficiency gains, or b) a way of organising service delivery: from no integration, to partial integration or full integration [30,31]. For both of these approaches, integration is a process that "occurs at different levels of the health system (regional, district, health facility) and in relation to key health system functions namely governance, financing, planning, service delivery, monitoring and evaluation, demand generation" [27,32]. Analysis of integrated health care requires a good understanding of health care services and health systems components and functions, and integration must be viewed as a process that accommodates complex agendas and issues as was concluded in a study of postnatal care integration into PMTCT in Swaziland [31,33].
In terms of implementation practices, PMTCT interventions may be carried out in one or more of the following health services: (1) antenatal clinic, (2) delivery/obstetric/ labour ward care, (3) postnatal care, (4) neonatal/new born care, (5) paediatric/infant care, (6) nutritional programs, (7) HIV testing and support centres, (8) HIV treatment centres, (9) reproductive/gynaecological services, (10) sexually transmitted infection clinics, (11) family planning, (12) primary health care which in many settings may be largely focused on conditions such as malaria, acute respiratory infection, diarrhoea and malnutrition, (13) Emergency care, (14) Tuberculosis clinics, (15) Malaria clinics (in areas with high burdens of malaria and HIV), (16) Immunisation, (17) or other service [27]. It is worth underlining the close ties between HIV including PMTCT and tuberculosis or/and malaria in terms of coinfection, prevalence, targeted groups and global funding initiatives [34][35][36]. This variety of potential entry points in the context of the discussion of potential synergies suggests how, and how much, the integration or non-integration of PMTCT could contribute to overall strengthening and integration or weakening and fragmentation of services and systems. In any case, the outcomes depend on how actors collaborate between themselves at health system level, and with the community or individuals seeking health care. A constructive cooperation is encouraged [30,37] to make integration work and minimise its potential negative effects.
One example that demonstrates the need for cooperative efforts to avoid negative effects of a vertical program on health systems and communities relates to the universal child immunisation goals. A study in six countries in Africa and Asia documented how a top-down model in immunisation interventions ended up creating conflicts between local demand and targets of immunisation policy, leading the authors to argue for more intersectoral collaboration if a specific programme is delivered and managed in a vertical way [28]. This example is one among many that supports calls for shifting from a vertical view of PMTCT programs of HIV prevention and treatment towards a horizontal focus on maternal health care and other health care services [14,38].
Unfortunately, the lessons that might be learned from immunisation programs and applied to thinking about PMTCT and health systems have not yet been taken fully into account.
Building on this overview of PMTCT and health system integration, this review sought to describe: (1) the effects of PMTCT on health care services and systems in SSA (2) the integration of PMTCT within broader programs and health systems and extent to which it occurred, and a related question (3) whether health systems as a whole have been strengthened or weakened in countries of SSA.

Objectives
This systematically conducted narrative review [39] was based on the following two objectives: 1. To describe the impact of PMTCT on health care services and systems in SSA and assess whether the PMTCT has strengthened or weakened health systems in SSA To describe the integration of PMTCT and its extent within broader programs and health systems Search 1: (prevention of mother to child transmission or MTCT or PMTCT or (Transmission* or spread*)) adj3 (Mother* adj3 child*)) or HIV or AIDS or acquired immune deficiency syndrome The identified papers through these three combined searches were all exported to Endnote bibliographic software [40].
The search strategy did not explicitly require the terms "impact" and "integration" because we felt that some studies might address "impact" and "integration" indirectly, through services description. We defined impact based on the existence of one or more areas of public health action framework as demonstrated through the five aspects of Frieden's 2010 pyramid that underlies intervention with potential health impact [41]. These five aspects are: 1) socioeconomic factors, 2) changing the context to make individual's default decision healthy, 3) long lasting protective interventions, 4) clinical care, 5) counselling and education [41]. Integration was defined based on the concept of integrated care as it was above explained.

Inclusion and exclusion
We designed the review inclusion and exclusion criteria for the purpose of retaining all studies pertaining to comprehensive PMTCT components throughout health systems in SSA as discussed in the WHO PMTCT strategy [6]. We looked for the papers that examined this strategy in terms of PMTCT's positive and negative effects on other health care services or vice versa, and integration of PMTCT programs with other health care services in the paper's title, results and discussion.
The abstracts of pertinent papers were then retrieved following these inclusion criteria before selecting the full articles: (i) Papers -Research articles published in peer-reviewed scientific journals, grey literature and commentaries dealing with PMTCT in pregnant women in SSA were accepted for inclusion. (ii) Participants -Women at risk of transmitting HIV infection to their children. This could include pregnant women or those at risk of pregnancy and their children, regardless of HIV status. (iii) Interventions -All interventions to prevent or reduce HIV MTCT, including but not limited to strategies for antiretroviral therapy and replacement feeding. PMTCT collaboration with other health care services especially maternal and child health (MCH) were included. The following types of articles were excluded: (i) Studies focusing on countries other than SSA countries, (ii) Studies focusing on general HIV/AIDS prevention or other health care services without reference to PMTCT and (iii) Editorials or commentaries generally describing the PMTCT programs on one or more pre-specified healthcare services without studying its effects or integration.

Data extraction and analysis
All three reviewers agreed on search strategy and inclusion and exclusion criteria. The initial database was created from the compiled electronic searches by one reviewer. All citations were firstly screened by title and abstract and duplicate citations were eliminated. The full texts of potentially eligible papers were then independently obtained for further screening. After resolving differences in data extraction or interpretation through consensual discussions based on the above stated inclusion and exclusion criteria, the final retrieval of papers was conducted. The following study characteristics and data were extracted from included papers: authors, year of publication and country of study, study types or designs, paper's focus interface with PMTCT, brief topic of investigation and main results. The findings were organised according to the objectives and referred to as 'themes' pertaining to each of two review objectives.
Throughout the whole selection process, the impact of PMTCT programs on other health care services or vice versa was reported in order to describe to how and to which extent it occurred. Integration of PMTCT programs with other health care services at different levels was also described across all included papers. Two guidelines on global scale-up of the PMTCT were also retrieved.

Results
We identified 6223 citations and ultimately retained 57 articles, in addition to the PMTCT guidelines mentioned above. These two guidelines are: 1) The global plan for elimination of new HIV infections among children by 2015 and keeping their mothers alive, and 2) The PMTCT strategic vision 2010-2015: preventing mother-to-child transmission of HIV to reach the United Nations General Assembly Special Session (UNGASS) and Millennium Development Goals [6,42,43]. The study selection process is presented in the form of an adapted PRISMA flow-diagram ( Fig. 1) and the retained articles are summarised in Table 2. Table 2 describes the included studies in which different types of research designs and data collection methods were used. Two papers were categorised as policy-evaluative (one that evaluates PMTCT policies and one that assesses various national strategic plans in 20 countries in Eastern and Southern Africa). One descriptive study presented a health program. Eighteen reviews, nine qualitative studies, thirteen quantitative studies, and eight mixed methods studies were selected along with six controlled trials and one quasi-experimental study. In most of the studies, hospital and clinical records were reviewed and individual or group interviews with women and staff members involved in health care and PMTCT services were conducted by the researchers       Risk factors for low PMTCT service uptake Strengthened linkages of referral-system and between communities and health facilities can address factors to low PMTCT service uptake.
*The paper described a health program in presentation form **National strategic plans in 20 countries were assessed in this study ***The paper evaluated policy issues and commented on them among the selected areas/sites and participants. Key informants such as policy makers, district health workers, academicians, implementing partners and persons living with HIV were also contacted in the research process in most of selected articles. Many programs and services were classified as fully vertical while a few others were perceived as either semi or fully integrated based on the extent of verticality or horizontality of PMTCT programs [14,44]. Thirty-two articles recorded the positive and negative effects of PMTCT on other health care services as a major theme in the article's title and/or discussion, out of fiftyseven [14,16,. Twenty-five articles examined PMTCT and its integration in health systems in SSA as a main theme [26,44,. Twenty-three papers out of 57 were classified as multinational as they covered two or more SSA countries [16, 44, 47, 52, 54-56, 59, 61, 64, 66, 68, 73, 74, 77, 81, 84, 85, 89, 95, 98], twenty examined our research themes in the Republic of South Africa (RSA) [26, 46, 48, 51, 57, 60, 70, 79, 80, 82, 83, 86, 87, 90-94, 96, 97], three in Kenya [53,69,78], two each for Malawi [62,71] and Tanzania [76,88] and one for each of Democratic Republic of Congo [49], Lesotho [72], Rwanda [67], Senegal [75], Swaziland [45], Uganda [63] and Zambia [65].
Integration as a theme was found in twelve papers on PMTCT integration in ANC, postnatal care (PNC) and child care [44,76,77,80,83,86,88,90,[95][96][97][98], five papers examining PMTCT and HIV services integration with tuberculosis screening and treatment [26,[91][92][93][94], six articles on integrating sexual and reproductive health (SRH) and family planning (FP) services to prevent unintended pregnancy and optimise maternal and infant health in ANC and PNC [75,78,79,82,85,89], one article on linking immunisation to HIV screening among children [87] and one on paediatric HIV [84]. We found literature examining both impact and integration of PMTCT programs as part of maternal and child health (MCH) services. MCH in this review covers antenatal, delivery and post-partum or post-natal care for both the mother and child. Other elements related to direct MCH services were included, notably social factors such as women's perceptions, community and male involvements in mother and child services, intimate partner violence and gender equality. The above-mentioned MCH components appeared in a total of twenty-one articles [14, 45, 47-49, 53-55, 57, 59, 60, 62, 63, 65-70, 72, 74].

Impact of PMTCT on other health care services
PMTCT has been promoted by WHO as a reliable solution to paediatric HIV. As a comprehensive approach developed based on the four components, the scaling up of PMTCT became a cornerstone of countries' HIV prevention, care and treatment programs [42]. As mentioned above, 32 of the 57 retained articles addressed the impact of PMTCT on other health care services. The major findings are the potential improvement of existing MCH services and to some extent increasing the availability, accessibility and utilisation of other PMTCT-related services such as prevention of unintended pregnancies, control of STI, immunisation, nutrition and vitamins supplementation [16,44,73,76,78,87,89]. PMTCT services also offer, to some extent, opportunities for screening tuberculosis among exposed infants [51] and may generally enhance diagnosis and treatment through collaboration efforts at both health care settings and community [26,91,92,94].
Several issues surrounding the PMTCT roll-out are reported to adversely impact PMTCT outcomes [14] and these include its process, services delivery, quality of supplies and tools, hiring and training of adequate staff, among others. However, conclusions regarding the impact of PMTCT must be cautious due to lack of detailed studies analysing such effects. In their attempt to quantify the impact of PMTCT programs on overall health systems in SSA, Nutman et al. [16] specifically assessed the existing knowledge and evaluated the PMTCT services impact beyond HIV transmission prevention. They also looked at how these programs contribute to the broader health outcomes. In the end, their systematic review of literature published up to 2011 found evidence of numerous beneficial synergies with specific health services but that there was insufficient evidence to draw any firm conclusions about the broader impacts of PMTCT on health outcomes or health systems. They report serious gaps regarding appropriate recording, data availability and information flow, and argue that these gaps might mislead decision-making, funding allocation and implementation of initiatives.

Health system strengthening or weakening
Health systems link people with promotive, preventive, curative, rehabilitative or palliative health services to address health problems. The effectiveness of this linkage depends on many factors, including those outside the health sector as well as within the various components of a health system [99,100]. We identified six articles explicitly or indirectly addressing the broader health system [44,46,59,86,90,91] while the remaining fifty-one focused internally, on one or more components of the building blocks and functioning of health systems and services.
Health care providers and funders across SSA have shown interest and engagement to make PMTCT services more and easily accessible to women and children [16,47,84,101] but their efforts have not eliminated many challenges underlying SSA health systems. Added to social problems, these health system issues create a situation in which it is difficult for women to actually utilise services during pregnancy and postpartum, even when these are offered free of charge [76,102,103]. Looking within the health systems themselves, PMTCT services have positively impacted MCH and adequately reduced the spread of HIV infection [16], but optimal outcomes occur where the health services are delivered in conducive working conditions, with adequately equipped facilities and committed management [104]. Such conditions are rare in SSA where the vertical transmission of HIV is the highest in the world. PMTCT and other HIV services in SSA depend on foreign funds estimated at billions of dollars but unfortunately the overall system and population-level results of such efforts and investments are seen as mixed [21].
Studies in this review pointed to these shortcomings and deficiencies in service delivery, especially for the components regarding the uptake of PMTCT programs such as antenatal HIV testing and receipt of test results, ARV prophylaxis and postnatal mother-infant follow up [105,106]. Campbell et al. [21] therefore argue for a strategy for (i) 'translating' intervention approaches into locally and culturally appropriate discourses and practices; (ii) building local capacity to sustain interventions once their funded period is over; and (iii) strengthening health systems in affected settings.
Analysing the effects of PMTCT on other health care services is often framed in terms of a debate over whether the roll-out of HIV/AIDS services including PTMTC did or did not strengthen the existing health systems and services. Studies in our review demonstrated the advantages of scaling up HIV services in terms of saving many lives, training health providers, and funding some key services like MCH, but the same findings highlighted the various shortcomings of HIV/AIDS interventions in health systems of low and middle-income countries and particularly in SSA [14,27,44,65,102,106]. Some encouraging results suggest positive impact of PMTCT programs on health systems [44] but also highlight the challenges that PMTCT and health systems did not meet; these unmet challenges are seen as negative effects mostly caused by poorly resourced settings facing a high disease burden [14]. These challenges include among others: continued high rates of home deliveries, shortages of personnel, inadequate supplies of test kits, varying distribution and availability of PMTCT service delivery points, lack of supplementary feeds for women who may opt for non-breast feeding for their infants, and logistical and social implications after testing HIV positive, such as a lack of spousal support and sometimes violence [14,103,106,107]. None of the retained papers showed or argued that PMTCT programs directly weaken health systems. In order to strengthen health systems, the implementation of PMTCT and disease-specific interventions needs more collaborative efforts to address structural, organisational, managerial and finance barriers.

Integration of PMTCT within broader programs and health systems
Even though some programs remain disease-specific, our review found that PMTCT programs and other health services mutually interact. PMTCT integration within general health care services in low and middle income countries has also been recommended to boost the utilisation of these interventions [27]. Some studies noted that under a PMTCT umbrella, the quality of other services in integration is also being closely monitored and improved [102,105]. PMTCT integration has for example, positively influenced maternal and child care services regarding service availability, accessibility and utilisation [44]. Evjen-Olsen et al. [76] suggested that maternal and neonatal health can be improved by integrating health care services, supporting integration of health systems rather than separately organising and managing different vertical and horizontal programs especially in developing countries [108]. The articles reviewed here also identified potential synergies between the integration or combination of PMTCT with specific health care activities outside of direct obstetrical and child care or MCH services, including sexually transmitted infection (STI) control and immunisation [16,71], sexual and reproductive health and family planning [56,58,61], nutrition [64], tuberculosis [51] and vitamin A supplementation [73]. The synergies are variously achieved in different contexts through progressive efforts, such as staff training and motivation, planning and evaluation of services, restructured management and financing among others. In terms of integration itself as a theme, the following synergies were examined by retained studies, apart from MCH: HIV/PMTCT services integration with tuberculosis screening and treatment [26,[91][92][93][94], with SRH, STIs and FP services [16,75,78,79,82,85,89] and with immunisation and HIV screening among children [87]. These linkages helped to increase and improve training of care providers, to review and enhance funding and implementation policies, to increase access and adherence to services, to reduce drug stock outs and improve basic infrastructures.
While identifying and describing the effects of PMTCT on other health services and health systems or the integration of PMTCT and its extent within broader programs and health systems, collaborations and involvements at different levels emerged to be crucial. Since health services are not only provided by the public sector, our search included other organisations offering PMTCT services and other actors involved in offering such services outside of the public sector. We identified examples of service delivery provided by NGOs [26], or delivered in refugee camps [88] while one paper focused on male partner involvement [60]. While studies examining public health services tended to focus on activities within clinics and services, these additional papers reveal that PMTCT activities shape interactions between the community members, social organisations and clinics offering the PMTCT services. The collaborations reported in some articles were directed to implementation or evaluation of initiatives against stigma, for the "normalisation" of HIV as just another other disease and to increase the accessibility of other social and support services to women living with HIV [6]. Selected articles stressed collaboration initiatives so that the PMTCT programs could become a model in "networking, nurturing relationships and bringing all available resources and agents to the table to find solutions and forge partnerships in order to procure all elements essential to a high quality, comprehensive, integrated program" [42].

Discussion
The review aimed to document the possible effects of PMTCT on health care services and systems and the integration of PMTCT within broader programs and health systems in the available literature in SSA. The analysis of available evidence regarding the two main aspects of this review, namely PMTCT impact and integration, supports a generally positive evaluation of positive synergies with MCH as well as other health services, suggests increasing partial integration of PMTCT within health systems, and offers inconclusive arguments on whether health systems as a whole have been strengthened or weakened by the PMTCT programs in countries of SSA. The discussion below addresses three issues that must be further studied and analysed in order to increase the likelihood that health systems will be demonstrably strengthened: the availability and quality of information, synergies and impacts within and beyond the health sector and the need to engage programmes and services beyond MCH, sexual and reproductive health, and TB.
Beyond the structural limitations on the positive impact that PMTCT programs might have on health systems in SSA, there is also a crucial continuing problem of accurate and timely information as highlighted by Theuring et al., (22), and a lack of rigorous research regarding the effects of PMTCT programs on health care systems or vice versa. While all included papers contributed to one or another dimension of the research questions, only two among 32 retained articles that recorded the impact of PMTCT, explicitly and directly focused on or addressed the impact of PMTCT on other health services and on health systems [14,16]. This is a very unusual scenario, considering the large body of available literature on HIV/AIDS interventions in SSA. It is worth repeating that Nutman et al. [16] specifically aimed to evaluate the impact of PMTCT services beyond transmission prevention and assessed the existing knowledge about such programs and how they contribute to broader health outcomes. Unfortunately, these research efforts were stymied by the weakness of health information systems, unreliable administrative and research data, and important evidence gaps. This research gap, coupled with under-or improper reporting practices, poses serious challenges to informed decision making, proper funding allocation and effective implementation.
Likewise, in their 2011 Cochrane review, Lorainne Tudor Car and colleagues did not find any study that evaluated the PMTCT interventions integration with other health care services to improve health outcomes [27]. In another study conducted in Swaziland, a country with a background of high, but stable HIV prevalence, an intervention with triangulated data of HIV testing and counselling, ART, PMTCT, and TB screening programs improved coverage and documented promising health outcomes. Even though the scale-up of this intervention was successful, the lack of data remained a non-negligible challenge throughout the study [109]. A situation like this is not an exception in SSA where health records are often incomplete or poorly managed. Without data storage and sharing, the combination of health interventions becomes more complicated and even impossible. This tendency can only be changed at national and regional levels through viable, coordinated investment and practice to monitor and to improve effects of major programs like PMTCT on other services and health systems or vice versa.
Synergies, collaborations and impacts on health systems: Positive contributions to a problem beyond the scope of PMTCT and the health sector The included papers provided moderate evidence of mutual benefits between the PMTCT services and the existing health care services and they recommended a close relationship and more integration in order to maximise the advantages of working together and to mitigate some of the challenges of controlling HIV, a lifelong condition, in the conditions facing SSA [8,14,27]. The available scientific studies and reports suggest the existence of both positive and negative effects generated by PMTCT on the maternal health services where they are mostly based and on child health services. The PMTCT services increase not only the accessibility but also the utilisation of antenatal and other MCH services [110]. In addition, the increased access to HIV testing and treatment are transforming HIV into a chronic illness. However, the lifelong management of HIV requires continued commitment and engagement of different actors such as the funders, the health care system, the patients, their families and as well as the community [4]. The efforts to manage the HIV pandemic as a chronic disease and treat the opportunistic infections or any other HIV associated conditions overwhelm SSA's struggling health care systems. The systems that depend on conditional foreign aid do not enjoy managerial autonomy and cannot consistently plan for their future interventions. PMTCT services within such systems cannot, on their own, strengthen health systems.
As seen in the results section, the failure of PMTCT to demonstrably strengthen health systems is associated and sustained by the underfunded and poorly managed health care settings [14]. HIV counselling and testing is not yet universal and many pregnant women come late to seek prenatal care and they then access all the services at the same time instead of progressing through the whole cascade. Lack of adequate resources and facilities (infrastructure, materials, and human resources among others) are major barriers to successful interventions in SSA where the operational and implementation problems arise from local contextual issues as well as the underlying poverty reflecting global economic forces.
The reality of plural health systems and multiple actors must also be addressed in research, policy, and practice. In SSA, the private for-profit sector and charitable and faith-based health care providers including mission hospitals and NGOs play an important role [26,91] and should be included in research and policy dialogue. This entails the widely advocated but in practice difficult coordination between well-funded international organisations, struggling national public health systems, NGOs and people living with HIV in search of treatment [111].
Beyond general statements of goals of collaboration and consideration of intervention context, there should be a clear strategy to strengthen local responses to HIV which are often ignored by the top-down style adopted in order to comply with requirements of "the global funding architecture" [112]. This responsiveness to local conditions and actors takes time and requires flexibility. In contrast, the "emergency" nature of much HIV intervention combined with accountability requirements to funders tend to align with a more directive and hence less responsive and collaborative approach. Constructive collaboration is not only advised but critically needed in order to achieve both PMTCT and health services goals. This concerns all models of health care delivery, whether vertical, horizontal or diagonal.
Beyond MCH, sexual and reproductive health and TB Malaria, severe anaemia, diarrheal diseases and acute respiratory diseases are some of the leading causes of death among women and children in many of SSA countries [14,113] including RSA, the source of most studies identified for this review. These are rarely, if at all addressed in PMTCT programs and their impacts on PMTCT or vice versa are not discussed in any study included in this review. This is possibly due to HIV interventions and research continuing to be seen as a separate issue from health conditions other than those directly related to sexual and reproductive health and TB.
In the cited systematic review of Lorainne Tudor Car et al. [27] on integrating PMTCT programs with other health services for preventing HIV infection and improving HIV outcomes in developing countries, the authors decided not to make any recommendation about the implementation of integrated PMTCT programs based on the fact that only one study met their selection criteria. It is logical to conceive and integrate the PMTCT programs within MCH services given the correlation of both services delivery but other services such as SRH or immunisation clinics as the point of entry to PMTCT programs need to be explored. In addition to publications for this review that recommend more research on the impact of PMTCT programs on health care services, the WHO, UNAIDS, UNICEF, UNFPA and PEPFAR calls to integrate the PMTCT and other disease stand-alone programs with other health care services [6,10,12,42] remain relevant.

Limitations
This a comprehensive and complex review that included a range of issues related to PMTCT impact and its integration on other health services and systems. Even though it was rigorously conducted, not all PMTCT implementation details could be reviewed in this single paper.
The regional HIV prevalence and other crucial factors, such as relationships with donors and international researchers as well as publication outputs, are highly variable across countries and regions in SSA. Most of the identified studies covered eastern and southern Africa. Central and western parts of Africa are less covered in publications, a situation that makes it difficult to generalise this review's findings to those regions. Studies reporting good service delivery and adherence may also not reflect the full impact of PMTCT when implemented at scale at a national level. Some areas within countries may also have been left out in the retrieved studies.
As this review sought articles specifically addressing PMTCT and health services and systems, it may have missed publications reflecting recent increased attention and funding for health system strengthening through HIV programmes, which did not specifically address PMTCT. These papers may yield insights into broader health system challenges and promising approaches to integration of other HIV programmes, potentially including PMTCT.

Conclusion
PMTCT improves maternal and child health through preventing the spread of HIV infection in SSA countries. There is evidence of a positive impact of PMTCT on primary care for mothers and children, beyond HIV. The provision of PMTCT services increases the availability, the accessibility and the utilisation of antenatal care and other health care services, especially when the intervention is linked to PMTCT programs as part of mainstream MCH services. However, this review also documented a large number of challenges involved both in implementing and in understanding the effects of PMTCT integration. Without robust information systems and rigorous and systematic research on the health system as a whole as well as on its various services and activities, evidence will continue to be fragmented and firm conclusions will continue to be impossible to draw.
While some vertical programs persist, PMTCT services are increasingly being integrated at different levels within routine health services and the health systems. Our study documented the challenges and weaknesses that face health care services and health systems in connection with PMTCT services. These range from structural, governance and resourcing challenges within and between countries, to weak information reporting systems and require more and better coordination and collaboration within and beyond HIV programs, directly related health services, communities, the plural health sector, and other sectors at national and global levels. These broad health system and social challenges cannot be solved by PMTCT interventions alone and there is a need of working together or collaborating with other sectors out of health system.