Facility-based delivery in the context of Zimbabwe’s HIV epidemic – missed opportunities for improving engagement with care: a community-based serosurvey

Although decreasing in many regions of the world, maternal and infant mortality rates are still far from the Millennium Development Goal targets for 2015. [1, 2] In 2013 almost 300,000 maternal deaths [1] and over 2 million early neonatal deaths occurred globally, [2] the majority in developing countries. With up to 40 % of maternal and newborn deaths and stillbirths occurring within 24 hours of birth, [3] skilled birth attendance has been strongly promoted. [4] Although there have been some successful initiatives to train existing traditional birth attendants in the community, [5, 6] in developing countries skilled birth attendance is largely synonymous with facility-based delivery. [7]

In addition to its role in maternal and neonatal health, facility-based delivery is recommended for the prevention of mother-to-child HIV transmission (PMTCT). Specifically, effective PMTCT requires the uptake and retention of pregnant women in a cascade of services including HIV testing, early uptake and adherence to antiretroviral (ARV) prophylaxis, safe obstetric practices and infant feeding counseling. [8] Vertical transmission of HIV during pregnancy, labor and breastfeeding can be reduced from 15 to 45 % to <1 % with appropriate and timely ARV-based interventions. [9] In the absence of ARVs, the greatest transmission risk occurs in the intrapartum period. [9] Home delivery has been cited as a contributing factor to limiting coverage and retention of mothers in PMTCT programs in sub-Saharan Africa. [10, 11] For example, a recent review found that home delivery is a key factor contributing to poor adherence to short-course ARV prophylaxis for PMTCT in sub-Saharan Africa. [12, 13] Facility-based delivery remains critical under Option B+ (the current WHO-recommended PMTCT guidelines whereby all HIV-infected pregnant women receive lifelong ART), [8] given the high loss to follow-up after initiating ART. [14]

Zimbabwe, where our study is based, has high maternal and infant mortality; the 2012 census reports 525 maternal deaths per 100,000 live births and 64 infant deaths per 1000 births. [15] Moreover, 12 % of pregnant women are HIV-positive [16] and at least 9 % of their infants become HIV-infected. [17] Only 65 % of births in Zimbabwe take place in health facilities according to the 2010–11 Zimbabwe Demographic Health Survey (ZDHS). [16] The 2010–11 ZDHS data also report that <5 % of women who deliver outside health facilities do so in the presence of a skilled attendant. [16] In the 1980s and 1990s traditional birth attendants were officially recognized and trained as skilled midwives; [18] however, since then their training has been actively discouraged. [19] Women from rural areas, who do not access antenatal care (ANC), who have high parity, with low levels of education, or who are poor are more likely to deliver at home. [16]

Due to the documented increased risk of non-adherence to ARVs and mother-to-child transmission associated with home-based delivery, [10‐13] delivering in a health facility is particularly critical for HIV-infected women and health staff are instructed to emphasize its importance during ANC. Hence, women who know they are HIV-infected should be more likely than their counterparts to deliver in a health facility because of the enhanced counseling they receive on the importance of facility-based delivery for safe delivery and PMTCT. However, little is known about whether learning one’s HIV status affects one’s decision to deliver in a health facility or not, and the limited existing evidence is mixed. [20, 21] This evidence gap is particularly relevant in the context of substantial technical and financial investments in national PMTCT programs in developing countries in recent years, including in Zimbabwe. Moreover, a better understanding is needed of the key points of contact with the healthcare system for women who do not deliver in health facilities, as these represent missed opportunities to inform women of the importance and benefits of facility-based delivery. Using representative survey data from Zimbabwe, [17, 22] here we examine: 1) the uptake of facility-based delivery and community-level variations in uptake in Zimbabwe in 2010–2011, 2) the association between HIV status and uptake of facility-based delivery, and 3) the prevalence of women who did not deliver in health facilities, depending on whether they accessed ANC and were tested for HIV.

We analyzed data from a 2012 baseline cross-sectional survey conducted to evaluate Zimbabwe’s accelerated national PMTCT program implemented in 2011. The objective of that evaluation was to assess the population-level impact of the PMTCT program on MTCT and HIV-free child survival at 9–18 months postpartum. [22] The methods have been published in detail elsewhere. [17, 23, 24]

In brief, infants (alive or deceased) born 9–18 months prior (henceforth ‘index babies’) and their mothers or caregivers (≥16 years old) were eligible for the community-based baseline survey. The infants’ age range was chosen to meet the objectives of the impact evaluation. [17] For this analysis, we excluded data regarding caregivers (n = 349, 3.9 % of 9018 participants) and only analyzed data on living biological mothers who were present at the time of the survey (n = 8662), as data collected about deceased (n = 55) and unavailable mothers (n = 294) did not include information about their place of delivery.

The survey was conducted in April-September 2012 in five of Zimbabwe’s ten provinces (Harare, Mashonaland West, Mashonaland Central, Manicaland, Matabeleland South). These regions include both major ethnic groups (i.e., Shona, Ndebele), some of the largest cities in Zimbabwe, and rural areas with higher and lower HIV prevalence. Study participants were identified using a two-stage stratified cluster design. Firstly, of the 699 health facilities offering PMTCT services in these five provinces, we randomly selected 157 facilities, proportionate to the number of facilities in each district. Secondly, we identified all eligible infants from the catchment areas of these 157 facilities and sampled a known fraction of them proportionate to the size of the target population in each catchment area. Eligible mother-infant pairs were identified based on information pooled from community health workers and immunization registers from selected and neighboring facilities (to identify women residing in sampled facilities who accessed services at facilities nearby). Further, those mothers identified using community health workers and immunization registers were asked to identify other infants in their neighborhood who were born in the previous two years.

We visited 42,995 households; 21,047 households (49 %) had at least one eligible mother-infant pair. In 571 households (1.3 %) we were not able to assess whether an eligible mother-infant pair lived there or not. Overall, we identified 21,205 eligible mother-infant pairs of which 9184 (43 %) were invited to participate in the survey. Of these, 9080 (98.9 %) mother-infant pairs consented to the questionnaire. After excluding corrupted files, questionnaires were available regarding 9018 (98.2 %) mother-infant pairs. We limited the analysis for this paper to 8662 living biological mothers who were present at the time of the survey.

In this paper we examined the uptake of facility-based delivery among women recruited in a large survey of mother-infant pairs conducted in Zimbabwe in 2012, one of the first studies to assess population-level estimates of HIV-free infant survival accounting for breastfeeding-related transmission using a community-based sample. [17, 22] We found that 77 % of mothers reported facility-based delivery, which varied considerably by community. The likelihood of facility-based delivery was not associated with self-reported maternal HIV status before delivery. Mothers who neither accessed ANC nor tested for HIV were the most likely to deliver at home (69 %) compared to other patterns of service utilization, however most home-based deliveries were among women who both accessed ANC and were tested for HIV infection. Although only 20 % of mothers who accessed ANC and reported testing HIV-negative delivered at home, these represented over two thirds of all home-based deliveries.

Our estimate of the overall uptake of facility-based delivery (77 %) is higher than the 2010–2011 ZDHS estimate (65 %). [16] Differences in sampling strategies between the two studies notwithstanding, this is likely indicative of the difference in survey timing and target populations i.e., our estimate included live births during the 9–18 months (2010–2011) prior to the survey in five provinces, while the ZDHS estimate included all live births to women in all ten Zimbabwean provinces in the five years (2005–2011) preceding the survey. Although our survey covered only half of the nation’s provinces, our estimate is likely a more accurate indicator of more recent uptake of facility-based delivery in Zimbabwe. The difference between the two estimates may suggest that the declining trend of deliveries in health facilities documented by ZDHS data (72 % in 1999, 68 % in 2005–2006, 65 % in 2010–2011) [16] has ended or reversed and uptake of facility-based delivery is now increasing.

Uptake of facility-based delivery was similar among HIV-infected and HIV-uninfected mothers, as also documented in Kenya. [31] However, in Rwanda 2002–2203 program data indicated that facility-based delivery in PMTCT programs is 2.7 times higher than in general population. [21] Given Zimbabwe’s substantial investment to retain HIV-infected women in care in recent years (e.g., approximately $USD 40 million in 2012), [32] and the counseling messages that strongly advise women who test HIV-positive during ANC to deliver in a facility, one might expect HIV-infected women (at least those who know their HIV status) to deliver in health facilities in larger numbers than HIV-negative women. At the same time, because they are expected to deliver in a facility, HIV-infected women who have not disclosed their status to community members may be reluctant to deliver in facilities to avoid disclosure, stigma and discrimination which continue to be important deterrents to uptake of HIV services Zimbabwe. [33] Non-disclosure of HIV status has been linked to low facility-based delivery, [31] poor uptake of PMTCT [31, 34‐36] and maternal and child health services [37] in developing country settings. Further, HIV-related stigma was associated with low uptake of facility-based delivery in Kenya. [20]

We find that women who reported they did not know their HIV status before delivery were significantly less likely to deliver in a health facility than those who reported knowing their status (44 vs. 80 %). The laboratory-confirmed HIV prevalence among mothers with unknown self-reported HIV status is 13.6 %, similar to the prevalence among all mothers in our sample (13 %). Assuming that women’s perceived risk of HIV infection is consistent to their actual HIV status, this suggests that the low uptake of facility-based delivery among women with unknown self-reported HIV status may not be an indication of perceived risk of testing HIV-positive, although available data do not allow us to further examine this hypothesis. Although women with unknown self-reported HIV status are the least likely to deliver in health facilities, particularly if they did not access ANC, they represent only 19 % of home-based deliveries overall.

The large majority of women who delivered at home accessed ANC: of the 2043 women who delivered at home included in the survey, 1387 accessed ANC and tested HIV-negative compared to only 216 who did not access ANC and did not test for HIV, suggesting that it should be possible to reach most women who are at risk of home-based delivery through their contact with health services. Hence, community and clinic-based staff need to continue to convey information about the importance and benefits of facility-based delivery for all women during pregnancy. Additionally, other approaches could further convince women to deliver in a health facility. Financial incentive schemes, [38, 39] community referral/ transport systems, [39] use of mobile technology, [39] birth preparedness and planning, [40] and male involvement are promising interventions for improving uptake of facility-based delivery, although further evaluations of their effect on maternal and infant mortality are needed.

In over half (57 %) of the communities, fewer than 80 % of recent mothers had delivered in a health facility. In addition, we found considerable variation in the community-level uptake of facility-based delivery, suggesting that there may be important community-specific factors in favor or against the practice of delivery in health facilities (e.g., social norms, religious norms), and/or facility-level factors including perceived quality of care among community members. Our survey collected limited information on community and facility characteristics, hence we cannot explore this hypothesis in detail. Future studies may want to explore community-level factors in more detail in order to better understand how these affect beliefs and practices related to childbearing.

The primary intention of this survey was not to assess mothers’ place of delivery. Consequently, various factors related to facility-based delivery were not examined in this study, and the place of delivery, healthcare utilization variables, as well as the self-reported HIV status before delivery were not verified against medical records. Moreover, these self-reported data relate to behaviors that participating mothers engaged in 9–18 months prior to the survey; hence, these data are subject to recall and social desirability bias. Nonetheless, the key variables included in our analyses relate to major events in the participants’ lives (e.g., birth of their child, finding out their HIV status), which are inherently memorable. Further, data were collected anonymously, which should minimize social desirability bias. Notably, the analysis excluded women who had died by the time of the survey, who may have been more likely to be HIV-infected. Although large-scale and representative, the survey was conducted in five of Zimbabwe’s ten provinces, allowing the possibility that uptake and patterns of facility-based delivery are substantially different in the non-surveyed provinces. Lastly, while we identified eligible mother-infant pairs using multiple approaches, isolated marginalized women may be underrepresented.

Further research is required to understand the ‘tipping point’ barriers and facilitators for service uptake faced by both HIV-negative and HIV-positive women to inform design of innovative interventions to increase demand and uptake of maternal health services and improve retention across the PMTCT continuum in the drive to virtual elimination of pediatric HIV. To meet its national goal of 85 % uptake of facility-based delivery by 2019, Zimbabwe aims to establish mothers’ waiting homes (residential facilities where pregnant women at risk of complications or who face distance as a barrier to uptake can await onset of labor) in all district hospitals by 2015. [41] Other interventions are underway through the Ministry of Health and Child Care and its partners to increase access and uptake to quality health services for women and children. Implementation science research is required to assess the impact of such efforts on uptake of facility-based delivery and on health outcomes of women and children, including PMTCT-related outcomes.

Our study found that 77 % of recent mothers in Zimbabwe delivered in a health facility. There was similar uptake of facility-based delivery regardless of maternal HIV status. The majority of women who delivered at home had previously accessed services suggesting missed opportunities for encouraging uptake of facility-based delivery. Women who failed to access ANC were the least likely to have tested for HIV and to deliver in a facility, suggesting that community health services need to intensify efforts to link these women to care.