Decentralizing and task sharing within the primary health system improved access and quality of ANC services in Amhara and Oromia regions: pre-post health facility data

Abstract

Background

Improved access to quality antenatal care (ANC) promotes healthy behaviors and early complication management, enhancing maternal and newborn outcomes. The Enhancing Nutrition and Antenatal Infection Treatment for Maternal and Child Health (ENAT) intervention in Ethiopia aimed to increase newborn birth weight by improving ANC utilization and quality. ENAT task shared and decentralized ANC services to facilitate early contact and point of care (POC) testing and management of maternal infections and anemia. This study assessed if task sharing and decentralization improved utilization and quality of ANC services at primary health care facilities.

Methods

The study assessed changes in ANC coverage and quality before and after the ENAT intervention, in 65 health centers and 303 health posts across Amhara and Oromia regions of Ethiopia. The intervention task shared ANC services at health centers (Augst 2018-January 2021) and then decentralized services to health posts (February 2021-February 2022). Using descriptive analyses, this study assessed, between baseline and endline, changes in coverage in the proportion of women who: enrolled early into ANC; had at least one ANC contact; and had four or more ANC contacts. Similarly, the study separately assessed each quality indicator through changes in the proportion of iron and folic acid supplementation, and deworming tablets, and POC testing and treatment for anemia, syphilis, and asymptomatic bacteriuria at baseline, phase I and phase II.

Results

ANC utilization and quality improved in the 368 sites. Between baseline and endline the proportion of women having an ANC contact before 16 weeks of gestational age increased from 6 to 37%, while those receiving at least one ANC contact and four or more ANC contacts increased from 74 to 91% and 45–57%, respectively. Iron and folic acid supplementation and deworming increased from 44 to 97% and from 44 to 79%, respectively. In the final 12 months of the intervention, 87%, 80%, and 87% of pregnant women attending ANC received POC testing for anemia, syphilis, and asymptomatic bacteriuria, respectively.

Conclusions

Our findings suggest that bringing ANC services closer to communities can increase early ANC contact and enhance the coverage and quality of services, leading to better maternal and newborn health outcomes.

Background

Antenatal care (ANC) is essential for improving maternal and fetal health by creating an opportunity to provide high-impact interventions during pregnancy [1]. However, access to quality ANC services is often limited in low resource settings [2, 3]. Poor quality ANC can lead to poor health outcomes for the newborn [4, 5]. Evidence has shown that newborns of mothers with inadequate prenatal care are at a greater risk of low birth weight compared to newborns of mothers with adequate pre-natal care [6, 7].

Trends in ANC uptake have shown improvements over the years. Coverage of at least one ANC contact and four or more contacts increased in low- and middle-income increased between 2011 and 2017 [8]. The 2019 Ethiopian demographic and health survey (EDHS) reported that 74% of women received at least one ANC contact, 43% received four or more ANC contacts and 48% gave birth in health facilities with skilled attendants [9]. However, tracking and engaging women in the first trimester (before 16 weeks of gestation) has been challenging. The 2016 EDHS data indicates only 20% of pregnant women started ANC services before 16 weeks of gestation [10].

In addition to geographical and economic barriers, distance, lack of resources, and poor attitudes of health care providers were identified as major barriers that hinder ANC uptake among women [11,12,13]. Even when women do attend ANC services, centralization of critical lab services for the diagnosis and treatment for maternal infection results in low coverage of these services and therefore low quality of ANC for most women [14]. WHO guideline recommends 49 interventions as essential components of antenatal care including: tetanus toxoid injections, screening and treating asymptomatic bacteriuria, anemia and syphilis; supplementation of iron and folic acid, and nutritional status screening and counseling [15]. According to the EDHS 2016 report, less than half of those who attended ANC received tetanus toxoid injections, were told about of pregnancy danger signs, and were tested for syphilis [10]. Furthermore, only 5% of the women who attended ANC took iron tablets for 90 days or more during their most recent pregnancy.

The Ethiopia Health Sector Transformation Plan II (2019/20-2024/25) has identified ANC as a priority service to prevent poor maternal and newborn outcomes [16]. One of the key focus areas to increase access to antenatal care is the provision of high-quality, equitable care. Furthermore, as part of global guidelines and the government’s effort to improve maternal and newborn health outcomes, pregnant women are now recommended to have eight contacts and at least one ultrasound service before 24 weeks of gestation [1].

The proximity of healthcare facilities to homes, their optimal management, infrastructure, and preparedness to offer ANC and community engagement are key factors in increasing the uptake and quality of ANC services [17,18,19]. The Enhancing Nutrition and Antenatal Infection Treatment for Maternal and Child Health (ENAT) project was designed to test a proof of concept aimed to increase newborn birth weight by increasing access to ANC services as well as improving the quality of ANC services. Specifically, the project intended to task share essential laboratory services at health centers so midwives can do point of care (POC) testing for pregnancy, hemoglobin test, urine and syphilis. In addition, ANC services were decentralized to health posts whereby health extension workers would provide the same POC testing to screen and manage infections and other major pregnancy related problems and provide improved nutritional counseling. ENAT intervention also aimed to improve the coordination between health posts and health centers, the availability of basic packages of ANC services, the structured mentoring to of health extension workers at the health posts and demand for ANC services in the community.

This paper aimed to assess whether the task sharing of essential laboratory services at health centers, decentralizing ANC services to health posts, enhancing the coordination between health posts and health centers, and structured mentoring to health extension workers at the health posts improved the ANC services. Specifically, the study assessed whether these measures increased coverage of early ANC contacts, number of ANC contacts and quality of ANC services.

Methods

Setting

The Ethiopian health system at the district level is comprised of several primary health care units serving approximately an average of one-hundred thousand people. Each primary health care unit is comprised of one health center and approximately five health posts and their district level hospital. Providers at health centers are nurses, midwives and public health officers who provide prevention, promotion and curative services. With respect to maternal and newborn care services relevant to this study, health center staff provide antenatal, delivery, and postnatal care as well as postpartum family planning. Health posts are run by two or more health extension workers (HEWs) who provide preventative and curative services under the health extension program. ENAT was implemented by Jhpiego and was jointly funded by Children’s Investment Fund Foundation (CIFF) and The ELMA Foundation. This paper is part of a cluster randomized controlled trial that aimed to look at the effectiveness of ENAT intervention on improving birth weight [20].

Study sites

To select implementation districts and facilities, obtain buy-in and finetune the work plan and activities, different consultative meetings were conducted with the Ministry of Health, Regional Health Bureaus and other lower-level health structures. These health centers were selected by the regional health bureaus as they were priority areas which had suboptimal maternal, newborn and child health indicators. The intervention area covered a catchment population of 2,153,014 and with an expected number of 73,462 pregnant women.

ENAT intervention

ENAT intervention was implemented in two phases. Phase I of the intervention was implemented from August 2018 to January 2021, in 65 health centers across 11 districts within three zones that were in Amhara and Oromia, two of the largest regions. Phase II of the intervention was implemented between February 2021 to February 2022, in 303 health posts within the catchment area of the 65 Phase I health centers. Phase I focused on strengthening ANC services at the health centers including task sharing of services. Phase II decentralized comprehensive ANC services to the health posts level while maintaining support for health centers.

Phase I of ENAT intervention

To understand the general status of the primary health care units, implementers of the intervention first conducted a Service Availability and Readiness Assessment and a Training Needs Assessment in 46 of the 65 health centers and 156 of the 303 health posts. Informed by training needs assessment, different job aids for ANC and POC testing were developed. Capacity building training was organized. More specifically, to improve the content and quality of ANC services, ENAT implemented the following:

• Clinical skills strengthening training was provided to midwives, nurses, and laboratory technicians on how to provide evidence-based quality ANC services, including ANC skills update and POC testing for asymptomatic bacteriuria, syphilis, hemoglobin and pregnancy. The intervention also strengthened the capacity of healthcare providers in the assessment, classification, and treatment of anemia, syphilis, and asymptomatic bacteriuria. The referral process for identified cases to higher-level facilities was also enhanced by introducing a standardized two-way referral system.

• Provision of essential equipment and supplies: health centers were supplied with necessary urine dipsticks, syphilis tests, hemoglobin tests, and essential ANC equipment, including a blood pressure apparatus and weighing scales. These were distributed through the public health system. Additionally, supply stock management was monitored through continuous follow-up.

• Improving nutrition intervention: Integrated with post-training follow up and supportive supervision, ENAT project field staff provided a need based on site coaching and mentoring on screening pregnant women for malnutrition, counseling on nutrition, and provision of deworming and iron and folic acid tablets. Additionally, essential job aids for nutritional counseling and gestational weight monitoring were developed and distributed.

• Mentoring and supportive supervision: To ensure the quality of care, mentoring and supervision was conducted every month for a sample of facilities and quarterly for all facilities. In Phase I, mentorship was provided directly by project staff who were nurses, midwives, and health officers not employed by the health system. The staff received Trainer of Trainer training on ANC, point-of-care testing, nutritional counseling and mentorship skills. Moreover, joint supportive monitoring was carried out semi-annually in selected facilities in collaboration with the public sector staff from different levels.

• One day training was also organized for midwives, HEWs and district level coordinators on ENAT intervention objectives, data capturing tools and data quality assurance techniques. Central and regional staff periodically visited implementation facilities to coach and monitor data management and quality assurance procedures. Data completeness, accuracy and consistency were assessed during each visit.

Phase II of ENAT the intervention

Although there were some improvements, the Phase I intervention approach had limitations. Phase I primarily focused on health centers with minimal interaction with satellite health posts. This approach didn’t meaningfully strengthen the continuity of care within the primary health care unit, as ANC services are provided at both health center and health posts. Thus, phase II, besides strengthening activities at the health center level, aimed to decentralize ANC services to health posts.

Health posts were divided into three categories. Category one health posts had to serve a large catchment population, have two or more HEWs with at least one HEW having a level four training (diploma level certification), and have space to conduct POC testing. Category one health posts provided a full package of ANC, including POC testing except for HIV and hepatitis. Category two health posts only required the presence of two HEWs and provided POC testing via an outreach model by health center staff who were trained to support the HEWs in doing POC testing. The rest were considered category three health posts and were supplied with Human Chorionic Gonadotropin tests and otherwise continued providing preventive and promotive services and linking pregnant women with health facilities providing comprehensive ANC service. Pregnancy tests were in included in Phase II due to the minimal improvement in early ANC enrollment seen in Phase I. Community level pregnancy testing in Phase II was intended to increase ANC enrollment among pregnant women < 16 weeks of gestational age.

For phase II, training materials and job aids were adapted to introduce the POC tests at the health post level and HEWs were provided with capacity building training and missing essential supplies and commodities for ANC services. For mentorship, Phase II utilized project staff as well as midwives, lab technicians, and reproductive health officers from health centers and district health offices. Midwives and laboratory technicians provided on-site skills transfer on identified gaps on POC testing and provision of ANC services to HEWs. Through onsite skill transfer, midwives and lab technicians built the capacity of HEWs on assessing, classifying and referring women found to be anemic and positive for asymptomatic bacteriuria and Syphilis. Further, to improve demand generation and revitalize community-level healthcare promotion, the intervention provided orientation to community and religious leaders on the importance of ANC for maternal and newborn health. A one-day demand generation orientation was also provided for a total of 395 participants, which included HEWs, health extension program focal persons and the volunteer woman development army leaders. The intervention also revitalized the volunteer Woman Development Army Leader’s system to identify pregnant women and encourage early ANC through home visits. To enhance pregnant women’s knowledge and decision-making, the already existing pregnant women’s conferences were leveraged. A key lesson from Phase I was that early ANC enrollment showed only a slight improvement compared to baseline data. To address this, HEWs began offering home pregnancy tests to enhance early identification of pregnant women and link them to the health system.

In phase I and Phase II

During both phases of the intervention, post-training follow-up, routine integrated supportive supervision and mentorship, and data quality assurance were conducted to all health centers and health posts. Periodic review meetings were also conducted with district, zonal health offices and program staff. Details of ENAT intervention activities are shown in Table 1.

Table 1 Summary of key activities accomplished by ENAT intervention in both phases during 2018–2022

Data collection

The pre intervention data used for baseline comparisons were obtained from existing records within the public sector health system, specifically from the District Health Information Software 2 (DHIS2) database. Retrospective baseline data for the previous 12 months (August 2017-July 2018) were collected using the tools adapted from the Service Availability and Readiness Assessment tool. These retrospective baseline data were assessed to provide a reference point for evaluating changes in key ANC service coverage indicators across the intervention phases. Data were also collected at the end of Phase I which lasted for two and a half years (August 2018 to January 2021) and at the end of Phase II which lasted one year (February 2021 to February 2022). Data for Phase I and Phase II were collected through health management information systems which were accessed from web-based data collection tools (DHIS2).

Data not collected through the routine health information system was collected as part of the Lot Quality Assurance Survey and Routine Data Quality Assessment, which were conducted at randomly selected intervention health centers biannually. A total of 1047 charts were randomly selected in 65 health centers to check the quality of the data and to also see, among recorded positive tests for syphilis, anemia and bacterial infections, the proportion that were treated.

Outcome measures

The increases in the coverage and quality of ANC indicators that were measured between baseline and endline (end of Phase II) at the primary health care catchment population are shown in Table 2.

Table 2 ANC coverage and quality outcome measurement indicators for the ENAT intervention

Analysis

The data collected by intervention staff on tablets was synced onto the central server, where the data was cleaned by the data manager. The study then assessed changes in early (< 16 weeks of gestational age) ANC, at least one ANC contacts, four or more ANC contacts, iron and folic acid supplementation, deworming tablets supplementation, screening and treatment for syphilis, anemia and asymptomatic bacteriuria, and nutritional screening (MUAC measurements). The study analysis involved descriptive comparisons of proportions across different time periods: baseline, Phase I, and Phase II of the intervention. ANC coverage indicators captured through the routine (DHIS2) data are aggregated, combining health center and health post data. Thus, the study was unable to capture health post data separately from health centers. The quality indicators were measured separately by comparing changes in the percentage of women counselled, screened for anemia, who received deworming tablets, tested for STI using point-of-care testing, and who tested positive for anemia and STI and were managed or referred across different time periods (baseline, Phase I, and Phase II). Through the Routine Data Quality Assessment study the study calculated the verification factor (VF) for a given indicator as VF = (Recounted # from facility registers - Reported # through system)/(Recounted # from facility registers*100). When the calculated VF was within ± 5% margin of error it was considered acceptable. When the VF was negative it implied that the given indicator was overreported while a positive VF implied that the indicator was underreported. A VF of 0 implied the indicators were correctly reported. We did not conduct formal statistical testing due to the study’s focus on descriptive trends rather than inferential statistics.

Result

Pregnancy test at health post level

During Phase I, pregnancy tests were not available at the health post level. In Phase II, a total of 7,154 women were tested for pregnancy. Of these, 2,647 (37%) were positive. Of those who had a positive pregnancy test, 1,671 (63%) were at or before 12 weeks of gestational age.

Antenatal care service coverage

Comparing baseline to Phase I period of the intervention, the proportion of pregnant women who had an ANC contact at gestational age of less than 16 weeks increased from 6% (1,914/31,892) to 23% (7819/33,881). However, the proportion of women who had at least one ANC contact decreased from 74% (23,600/31,892) at baseline to 64% (21,525/33,881) at the end of Phase I. Similarly, four or more ANC contacts decreased from 45% (14,352/31,892) to 43% (14,489/33,881) at the same time.

When comparing Phase I and Phase II (decentralization phase) of the intervention, the proportion of pregnant women who received at least one ANC contact increased from 64% (21,525/33,881) to 91% (38,896/42,743). The proportion of women who had four or more ANC contacts also increased from 43% (14,489/33,881) to 57% (24,340/42,743) (Fig. 1). Similar to Phase I, the proportion of women who had a first ANC contact at less than 16 weeks of gestational age increased from 23% (7,819/33,881) to 37% (15,925/42,743).

Fig. 1

Trends of ANC service uptake in ENAT intervention health facilities, 2018–2022

Pregnant women who received tablets for iron and folic acid and deworming

The proportion of pregnant women supplemented with iron and folic acid 90 + increased from 44% (10,384/23,600) at baseline to 97% (37,730/38,896) at the end of Phase II. The proportion of pregnant women were provided with deworming tablet also increased from 44% (10,384/23,600) at baseline to 79% (30,728/38,896) by the end of Phase II (Fig. 2).

Fig. 2

The proportion of pregnant women receiving iron and folic acid and deworming tablets at ENAT intervention facilities, 2018 to 2022

Pregnant women screened for syphilis, anemia, and asymptomatic bacteriuria using POC testing

Between baseline and endline (Phase II), out of the total women who had one ANC contact, the proportion of women screened for anemia increased from 56% (13,216/23,600) to 87% (33,840/38,896) and the proportion screened for syphilis increased from 42% (9912/23,600) to 80% (31,117/38,896). Testing for asymptomatic bacteriuria was not done prior to ENAT intervention. Hence, baseline coverage data was not available. Between end of Phase I and end of Phase II periods, the proportion of women tested for asymptomatic bacteriuria increased from 60% (12,934/21,525) to 87% (33,840/38,896) (Table 3).

Table 3 Pregnant women screened for anemia, syphilis and asymptomatic bacterial infection at ENAT intervention facilities, 2018–2022

Pregnant women treated for syphilis, anemia and asymptomatic bacteriuria

Based on the semiannual Lot Quality Assurance System on data collected from 1047 out of 5505 total charts, the proportion pregnant women diagnosed with syphilis who received treatment increased from 52% (387/744) by end of Phase I to 99% (661/668) by the end of Phase II. Similarly, between of Phase I and Phase II, the proportion of pregnant women diagnosed with anemia who were treated increased from 57% (50/87) to 89% (139/156) and the proportion diagnosed with asymptomatic bacteriuria who received treatment increased from 84% (131/156) to 97% (113/116).

Pregnant women who received nutritional screening (MUAC)

There were no baseline data for pregnant women who received nutritional screening. During Phase I and Phase II, the proportion of pregnant women screened for MUAC at their first ANC contact increased from 53% (17,079/32,224) to 97% (37,729/38,896), and the proportion screened at their fourth ANC contact increased from 17% (5478/32,224) to 92% (35,784/38,896) (Fig. 3).

Fig. 3

Proportion of pregnant women measured for MUAC at ENAT intervention facilities during ANC 1st and 4th visits, 2019 to 2022

Data quality improvement

The Routine Data Quality Assessment of indicators conducted from September to November 2020 and from January to April 2021 showed improvements from poor verification factors towards an acceptable range (± 5%). For example in one health center, the Routine Data Quality Assessment verification factors for the number of pregnant women who: had their first ANC contact at less than 16 weeks improved from − 14% to -3%; had four or more ANC contacts improved from − 71 to 3%; received iron and folic acid 90 + tablets improved from − 98 to 1%; and, received deworming tablets improved from − 91% to -9%.

Discussion

Improved access to high quality ANC services during pregnancy is important for the health of the mother and the newborn. This assessment showed that ENAT interventions increased the proportion of women who had at least one ANC contact, had their first contact at less than 16 weeks of gestational age and had four or more ANC contacts. Among pregnant women, the intervention also increased the proportion that received 90 plus iron and folic acid tablets, deworming tablets and POC testing for syphilis, anemia and asymptomatic bacteriuria. Treatment for these conditions and MUAC screening also increased over the course of the intervention.

Between baseline and end of Phase I of the intervention, the proportion of women who had their first ANC contact at less than 16 weeks of gestation improved by three-fold (7% vs. 23%) and by the end of Phase II one in three pregnant women (37%) had ANC contact at less than 16 weeks of gestation. The improvement in early contact could be due to the intervention’s effort to create demand for ANC services, provide free pregnancy tests at the health post level and improve referral linkages [21]. While this improvement is encouraging, more needs to be done. The current WHO and national guidelines recommend that women should ideally have their first contact at less than 12 weeks of gestational age [1].

The proportion of women that had at least one ANC contact and the proportion that had four or more ANC contacts did not show improvement during Phase I. This was likely since Phase I intervention was centered on health center. ENAT intervention only had light touch activities at health posts where ANC visits were also expected to take place closer to the community. The resurgence of the COVID-19 epidemic is also likely to have affected the uptake of ANC services during this period. Evidence from other studies have shown consistent results with our Phase I findings, highlighting that the lack of care provision for women within their own communities [17, 22], lack of community engagement [17, 23, 24], and occurrence of epidemics like COVID-19 within fragile health systems [25, 26] were associated with reduced uptake of ANC services.

To decentralize ANC services, the intervention implementer first assessed service availability and readiness, as well as training needs. The assessment identified gaps in knowledge, skills, equipment, and supplies. To address these, the intervention implementer provided HEWs with training in ANC skills, counseling techniques, and point-of-care testing (POCT). This training helped them to deliver better ANC services at the health post level. Our findings match other studies that Joseph C. et al. 2021, that show high-quality local services [19] and Downe S. et al. 2019, community engagement [17] improve ANC uptake.

To decentralize ANC services, we first assessed service availability and readiness, as well as training needs. The assessment identified gaps in knowledge, skills, equipment, and supplies. To address these, HEWs were provided with training in ANC skills, counseling techniques, and POC testing kits. This allowed them to deliver enhanced ANC services at the health post level. Studies have shown the uptake of antenatal care service increased with decentralization of services to lower levels [27, 28], community engagement [17, 24] and strengthened referral and feedback systems [29]. Indeed, with decentralization of services to health posts, ANC coverage in our study also increased. By the end of Phase II nine out of ten pregnant women had at least one ANC contact and over half of the pregnant women had four or more ANC contacts. Although the increase in four or more ANC contacts is promising, it is important that the WHO recommends women have 8 or more visits [15]. This was not assessed since at the time of the study, Ethiopia had not yet implemented this WHO recommendation.

The different aspects of the intervention, including introduction of POC testing, contributed to the increase of the proportion of pregnant women who were screened and treated for syphilis, anemia and asymptomatic bacteriuria. Early screening and management of sexually transmitted infections also improves birth outcome [30, 31]. In addition to introducing POC testing, the intervention also strengthened the capacity of healthcare providers in the assessment, classification, and treatment of anemia, syphilis, and asymptomatic bacteriuria. To ensure quality care, mentorship and coaching were integral components of the intervention. The referral process for identified cases to higher-level facilities was also enhanced by introducing a standardized two-way referral system. Additionally, supply stock management was improved through continuous follow-up.

Recent studies show that undernutrition among pregnant women in Ethiopia is high indicating the need for nutritional screening [32,33,34,35]. In this study, by the end of Phase II, almost all women had a nutritional screening through measurement of MUAC during their first and fourth ANC contacts.

The increased proportions of women getting early ANC contact, more contacts and better-quality ANC services has implications for improving maternal and newborn health outcomes. Indeed, the cluster randomized controlled study that measured the effect of ENAT intervention on newborn birth weight reported that risk of low birth weight newborns was significantly lower by 36% in intervention clusters as compared to control clusters [20].

To adapt to this intervention, the public health system needs to evaluate service readiness and training needs of primary health care system and provide POC testing kits and required trainings for lower levels of the health system. Creating awareness and demand for the available services in the community is also vital. After implementation, regular follow-ups, support, and mentorship through existing forums can help improve task sharing and decentralization of ANC services. Keeping track of POC testing kit availability is also crucial for maintaining quality ANC services.

In a health system where community facilities have strong infrastructure, adequate staffing, proximity to referral centers, and active community involvement, comprehensive ANC services can be decentralized. For facilities with inadequate staffing, limited space, and distance from referral centers, decentralization can occur through outreach support from midwives and laboratory technicians at referral facilities. However, facilities with poor staffing, weak infrastructure, and minimal community engagement need to address these challenges before offering comprehensive ANC.

This implementation employed a longitudinal data collection approach which ensured periodic follow up on outcome measures and generated strong evidence to address intervention objectives. Additionally, stringent data quality assurances techniques were employed to maintain high quality data for evidence generation.

The study has some limitations. This study could have been stronger if data were available from the sites that were not receiving the intervention. Both baseline and endline data were not collected directly from women and hence we are not able to measure other indicators that might affect coverage of ANC and other services (e.g., socioeconomic factors.), and underlying health conditions. During the COVID-19 pandemic, POC testing kits and supplies were interrupted temporarily and some facilities were shifted to COVID-19 treatment center, affecting the uptake and content of ANC services.

Conclusion

Decentralization of ANC services to lower level of health facilities through task sharing of laboratory services at health centers, introduction of point of care testing at health posts, improving nutritional screening, improving the availability of essential equipment and supplies, strengthening clinical skills of health care workers, providing supervision and mentorship and community engagement increased the coverage, early contact and quality of ANC services. Lower-level health care providers such as HEWs can provide essential packages of ANC service if capacitated, supplied with essential Point of Care testing and given appropriate and timely coached and mentored. Having early and more ANC contact provides an opportunity to identify problems early and provide high-impact interventions that can improve the health outcomes of mothers and their newborns.