A prospective observational description of frequency and timing of antenatal care attendance and coverage of selected interventions from sites in Argentina, Guatemala, India, Kenya, Pakistan and Zambia

The World Health Organization (WHO) recognizes the importance of antenatal care (ANC) within a continuum of reproductive and maternal-newborn care. WHO recommends early identification of pregnancy and at least four ANC visits, starting prior to 14 weeks gestation, during which a prescribed package of preventative, screening, and educational interventions are delivered [1]. Unfortunately, such a consistent, integrated continuum of comprehensive care—although defined on paper, and codified in many national health policies—often does not translate into actual practice in some resource-limited settings [2]. For example, while it is widely acknowledged that ANC is most effective if it is initiated early and consistently throughout pregnancy[3], many pregnant women in resource-limited settings (1) initiate ANC late in pregnancy (e.g., during the 2^nd or 3^rd trimester) and (2) do not receive the WHO-defined minimum of at least 4 ANC visits. In addition to the frequency and timing of ANC visits, the quality of ANC care, as assessed by comprehensive, effective coverage of key evidence-based elements of the WHO Focused Antenatal Care (FANC) package, is also of concern [4‐9]. Although the strength of the evidence-base for some components of the focused ANC package is variable, FANC as a bundled, integrated intervention can serve as an important pathway by which to increase the probability that women will chose facility-based births with skilled delivery attendants, versus home births with family members or unskilled attendants [10‐12]. It is important to delineate the gaps, barriers, and facilitators which impact access to comprehensive coverage of recommended ANC interventions, as this has profound implications for improving the entire continuum of maternal newborn child health (MNCH) care in low-resource settings.

Currently, the Global Network for Women’s and Children’s Health, funded by the National Institute of Child Health and Human Development (NICHD), is one of the largest global research networks focused exclusively on testing and generating evidence-based recommendations for improvement of maternal-child health in resource-limited settings [13]. Given the importance of an integrated ANC continuum toward positive health outcomes among our populations of mothers and babies, to assess the quality of ANC coverage among Global Network sites, and to identify relevant targets for future research and implementation initiatives, we sought to: (1) describe the quantity (i.e., number of visits; timing of ANC initiation) of ANC attendance and (2) quality of ANC attendance in terms of coverage of key preventative, screening, and birth preparedness components.

The Maternal Newborn Health Registry (MNHR) is a prospective, population-based birth and pregnancy outcomes registry that was started in 2009 at communities in seven Global Network sites, including: Argentina, Guatemala, India (Belgaum and Nagpur), Kenya, Pakistan, and Zambia. A detailed description of the overall purpose, methods, and data collection techniques for the Global Network for Women’s and Children’s Health Research Maternal Newborn Health MNHR is provided elsewhere [13‐15].

For the current study, MNHR data from 106 clusters for the period January 1, 2010 – December 31, 2013 were analyzed for variables related to the patterns of ANC (e.g., number and timing of ANC visits; where, and from whom, ANC was obtained) and coverage of selected elements of the WHO FANC package.

Data were collected and entered at each study site and transmitted through secure methods to a central data coordinating center at Research Triangle Institute International (RTI). Data edits were performed at each study site and then centrally. Descriptive statistics were generated overall, by global region (sub-Saharan Africa, South Asia, and Latin America), and for each individual site. In order to investigate the numbers and percentages of women (overall, by region, and by site) who reported having components of a birth/emergency preparedness plan, we also explored a sub-set of data which were collected during the course of a trial for implementation of community-based Emergency Obstetric and Neonatal Care (EmONC) [16]. Because the EmONC study intervention may have influenced results from the intervention clusters, for the purposes of the current study, we report the results from the women enrolled only in the control clusters. All analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA).

In order to present a more comprehensive assessment of the quality/quantity of ANC by site, we developed a composite variable composed of 15 ANC indicators. The composite variable allows a comparison of how each Global Network site complied with a variety of ANC indicators, which, when assessed collectively, provide an indication, within and between sites, of the rate of coverage for bundled ANC service delivery. For each indicator we arbitrarily grouped compliances and categorized these as good, fair or poor. As an example, for having at least one ANC visit, good was classified as >95%, fair as 85-94% and poor as <85%. Good, fair and poor for tetanus toxoid, provision of iron and vitamins used the same percentage categories. For all the other indicators, good was classified as >75%, fair as 50-75% and poor<50%. These included: having at least four ANC visits, syphilis testing; HIV testing; anemia testing; maternal height measured; maternal weight measured; birth plan developed; transport plan developed; birth attendant identified, and identified birth attendant present at delivery. We then simply compiled the number of good, fair and poor answers for each site to develop a figure to illustrate these metrics. Because the Argentina site expected virtually all women to deliver in a hospital, we coded the categories of birth plan developed; transport plan developed; birth attendant identified, and identified birth attendant present at delivery as good for this site.

The appropriate Institutional Review Boards and Ethics Research Committees of the participating institutions and the Ministries of Health of the respective countries approved the MNHR. Prior to initiation of the study, approval was sought from the participating communities through sensitization meetings. Individual informed consent for study participation is requested from each study participant. No monetary reimbursements are provided to study participants nor to the communities participating in the study. A Data Monitoring Committee, appointed by the NICHD, oversees and reviews the study at annual meetings.

Data from a total of 269,710 deliveries (excludes women who were lost-to-follow-up prior to delivery; miscarried; or underwent medical termination of pregnancy) were used to explore the frequency of ANC (Figure 1). By global region, 23% of these births occurred in African sites, 62% in Asian sites and 15% in Latin American sites.

Number of ANC visits (Table 1): Ninety-six percent of women reported at least one ANC visit during their pregnancy. In the Nagpur and Belgaum, India sites, respectively, 100% and 99.9% of women reported at least one antenatal encounter. High rates of at least one ANC visit were also reported among pregnant women in the sites in Zambia (99%), Guatemala (98%), Kenya (97%), and Argentina (95%). The Pakistan site reported the lowest percentage (84%) of women with at least one ANC encounter.

Across all sites, 53% of women reported having at least four antenatal encounters. The Nagpur (81%), Guatemala (75%) and Argentina (66%) sites reported the highest rate of pregnant women who accessed ANC at least 4 times, with sites in Belgaum, India following at 61%. African sites (Zambia, 22%; Kenya, 43%) and the Pakistan site (24%) reported the lowest rates of pregnant women with at least 4 ANC encounters.

Timing of ANC access (Figure 2): Pregnant women from the two Indian sites were more likely to initiate ANC during the first trimester (63% for Belgaum and 78% for Nagpur, respectively), and reported the lowest percentages of initial ANC visits during the third trimester (6% and 3%). Sites in Kenya and Zambia reported the lowest initiation of ANC during the first trimester, at only 4% and 8%, respectively. The Pakistan (43%) and Kenya (37%) sites reported the highest number of women who had their first ANC encounter during the third trimester of pregnancy.

Given the high percentage of women who initiated ANC during the first trimester in the Indian sites as compared to the other sites, we investigated the overall timing of ANC initiation with and without the India data included. Overall, with the India sites included, across all Global Network sites, 43% of women began ANC during the first trimester, 40% during the second trimester, and 18% during the third trimester of pregnancy. With the Indian data excluded, among women in sites in Argentina, Guatemala, Kenya, Pakistan, and Zambia, there was a marked shift toward first ANC visits occurring in the second (50%) and third (29%) trimesters.

Location and provider where ANC obtained (Table 1): A government hospital was the most commonly reported location where women across the entire study obtained ANC (44%), particularly in the Kenyan (83%) and Argentinian sites (81%). In the Zambian site, women received their ANC from either a government hospital (29%) or clinic (32%), while 35% received their care from “others.” Women from the Belgaum, India site reported frequent use of both government (48%) and private hospitals (27%); in the Nagpur site, women most frequently selected government hospitals (45%) and health workers (35%). The most frequent location where ANC was obtained in the Pakistan site was from private clinics (73%). In the Guatemalan site, 38% of women reported receiving ANC from a traditional birth attendant (TBA) and 29% from a government clinic.

Comprehensive ANC confers both independent and collective benefits for mothers and babies. When incorporated as part of an integrated continuum of MNCH services which also includes skilled birth attendance and postpartum care, ANC coverage can contribute to reduction in rates of stillbirth, maternal, and newborn mortality [68]. It is crucial, therefore, to identify gaps in access and coverage, in order to develop targeted strategies by which to ameliorate them.

Results from our study confirm that a crucial barrier to implementation of comprehensive ANC in our Network, particularly at sites in sub-Saharan Africa and Pakistan, is late initiation of care. Thus, in order to increase rates of early initiation of ANC and improve the potential for coverage of key interventions, stakeholders may want to consider implementing strategies for active pregnancy case-finding and explore avenues by which to achieve early, persistent linkage of pregnant women to timely and frequent ANC.

This is one of the largest prospective studies of which we know in which data were collected from a diverse number of global regions for multiple indicators related to the quantity and quality of ANC coverage. By assessing the patterns of ANC and coverage of selected key ANC indicators among a large, population-based sample of 269,710 women over 4 years, this study contributes important information regarding the quantity and quality of ANC services that women receive in rural regions of Africa, Asia, and Latin America. Our results provide investigators with high-quality observational data by which to further explore ANC in their own settings. We have identified key gaps for which stakeholders can develop targeted interventions by which to improve women’s access to an integrated continuum of respectful, comprehensive maternal-newborn and reproductive health services.

Reviewer reports for this article can be found in Additional file 1.