Coverage and factors associated with completion of continuum of care for maternal health in sub-Saharan Africa: a multicountry analysis

Although the Maternal Mortality Ratio (MMR) declined from 342 per 100,000 live births in 2000 to 211 deaths per 100,000 live births in 2017 globally, most sub-Saharan African (sSA) countries had difficulties in achieving the Millennium Development Goals (MDGs 4 and 5) [1, 2]. Similarly, the neonatal mortality rate declined by more than half (from 36.6 to 18 deaths per 1000 live births) between 1990 and 2017 globally [3]. MMRs and neonatal mortality rates, however, vary across regions. In 2018, 1 out of 37 neonates in sSA died compared to 1 in 333 in high income countries and MMR varied from 11 to 14 for high income regions to 511 to 652 per 100,000 live births for sSA [2‐4]. Building on the MDGs, the United Nations member states adopted the Sustainable Development Goals (SDGs) to reduce the global MMR to less than 70 per 100,000 live births by 2030, with no country exceeding twice the global MMR (140 per 100,000 live births) [5]. Globally, 2.9 million newborns and 265,000 mothers die annually due to complications from pregnancy, childbirth and postpartum period. Of these, more than half occur in sSA [6]. Moreover, annually, an estimated 2.6 million stillbirths occur globally and 98% of all these stillbirths occur in low and middle income countries with three-fourths occurring in South Asia and sSA [7, 8].

Many strategies have been implemented in an attempt to reduce maternal and child deaths [9]. Continuum of Care (CoC) is one of the strategies that gained recent recognition and is advocated by the World Health Organization (WHO) and other initiatives to reduce stillbirth, neonatal, child and maternal mortality [6, 10]. Many maternal and neonatal deaths are largely preventable in low income countries by expanding the CoC approach [11‐13].

Currently, CoC for maternal and newborn health (MNH) care is recommended as advantageous over each service provided separately to achieve the global target of ending preventable maternal, newborn and child deaths because each stage of the CoC builds on the success of the previous stages [6, 14, 15]. Lack of appropriate care at any stage of CoC leads to poor MNH outcomes [16].

Even though complete exposure to CoC has advantages over separate stages of care [17‐20], evidence from the globe suggests that completion of CoC for maternal health is very low [6, 15, 21‐31]. Many pregnant women (84.3%) received at least one ANC visit in sSA and South Asia, which varied from 43% in Ethiopia to 99% in Malawi. Only few women, however, utilize at least four ANC visits (37.9%) as previously recommended by WHO, had skilled birth attendance (28.3%) and received postnatal care (16.9%), indicating disparities in the coverage of each stage of maternal health care [32].

Several studies investigated associated factors of maternal health care utilization in sSA [33‐36]. Most were mainly country-specific, with a primary focus on Ethiopia [23, 27, 28, 37, 38], Ghana [6, 21] and Nigeria [25, 29]. Moreover, very few studies have used nationally representative data [25, 36, 37]. Their narrow geographic scope limits generalizability, particularly when countries have complex multi-ethnic settings. To the best of our knowledge, little evidence exists on the status of CoC and its associated factors on a sSA scale. Therefore, it is essential to generate updated information on factors associated with CoC in order to design effective strategies to increase CoC. Instead of looking into the different stages of maternal health care, this study investigated CoC and factors that influence women’s continuation in receiving care because CoC is the most underutilized care and critical intervention to improve maternal and newborn wellbeing [28, 39].

Completion of CoC occurred in only 56,284 (25.0%, 95% CI 20.5, 29.4) of the women in sSA. This was higher in Southern Africa with 51.5% (95% CI 37.5, 65.5) and much lower with approximately 20% in the other three regions. Large disparities of CoC were found across sSA countries with as low as 2.7% in Burundi and as high as 61.2% in South Africa. This finding is in line with pooled estimates of South Asian countries (24.5%) [32]. Our finding, however, is higher than the pooled estimates of a previous study in sSA (13.9%) [32]. The higher completion rate in this study might be attributed to the inclusion of recent and most DHS in sSA, while that study included only five East African countries and one West African country [32]. Possibly maternal health care utilization has improved substantially over time in several sSA countries [47]. Further, this discrepancy might be due to differences in timing of ANC initiation, since timing of ANC is one of the predictors for completion of the CoC [23, 48] and in our study, 39.1% of women initiated ANC within the first 12 weeks of gestation, higher than in the previous one (23.9%). Even though focused ANC includes at least four ANC visits, SBA and at least three postnatal care visits for all mothers and newborns, this comprehensive analysis of 32 DHS from sSA showed low proportions of women utilizing CoC [49, 50].

This study demonstrated that the inclusion of community-level variables was important in explaining the variations in the completion of CoC. Community-level variables such as region, residence, distance from health facilities and community education showed effects on the completion of maternal health care. Only 2.27% of the total variation remained unexplained after adjustments of individual and community level factors.

When maternal age increases, the odds of completing CoC increases, consistent with some previous studies [51, 52]. Some others, however, showed different effects [6, 22‐24, 27]. One study indicated that younger women were less likely to recognize pregnancy early [53]. It is also observed in that study, that 22.5% of women < 25 years were never married compared with 5.0% of women aged 25–34 years and 2.2% ≥ 35 years and most unintended pregnancies were among unmarried women. Younger women who are unmarried are not likely to disclose pregnancy to avoid potential social implications since premarital pregnancies are highly stigmatized [54]. Risks of maternal and fetal complications are, however, more likely among adolescent women and lower utilization of maternal health care by adolescents is particularly problematic [55, 56].

Consistent with studies in Ethiopia, Nigeria, Ghana, Chad, Gambia, Nepal, Pakistan, higher women’s education was significantly associated with increased odds of CoC [22‐25, 28, 30, 32, 36, 51, 52]. Women’s understanding of multiple dimensions of health and health knowledge increased with more education, leading women to seek greater access of acceptable maternal and new born care [57]. Moreover, educated women are more aware of health-protective information and have greater decision-making power and demand higher quality care and pay more attention to their health in order to ensure better health for themselves and their children [58].

Supported by studies conducted in Ethiopia, Chad, Pakistan and Nepal, women exposed to mass media were more likely to utilize CoC [22, 24, 27, 36, 46, 59, 60]. Mass media are important means of disseminating information concerning maternal health, increasing knowledge, attitude and behavior of women towards maternal health care utilization [61‐63].

As observed in previous studies, lower odds of completing CoC were observed among women in rural areas as compared to urban areas [24, 25, 28, 29, 31, 36, 37, 46, 59, 66]. Lack of adequate health facilities, a deficit in health care professionals, lack of infrastructure (e.g. road conditions, costs of transport), distance from health facilities are more prevalent in rural areas [22, 64]. Moreover, women residing in urban areas have better educational status and better chances of health information than rural women. Therefore, provision of maternal health care to rural women through home visits, outreach programs and mass media campaigns should be implemented to improve CoC [65].

Women with unintended pregnancies had lower odds of completing CoC as compared to planned pregnancies, similar to studies in Ethiopia, Ghana and Nepal [6, 25, 29, 30]. Women with intended pregnancies are more likely to detect pregnancy earlier and may be careful about their pregnancy status compared to those with unintended pregnancies [66]. Moreover, women with unintended pregnancies are less likely prepared emotionally and financially for the demands of pregnancy and childbearing and less likely to care for the baby [67].

Delayed initiation of ANC was negatively and significantly associated with utilization of CoC, in agreement with studies from Ethiopia, Gambia, and Japan [15, 23, 36, 48]. Early initiation of ANC gives the opportunity to discuss birth preparedness and complication readiness and helps women to receive health promotion and preventive care such as immunization against tetanus, nutrition counseling, prophylactic treatment of malaria and worms [68‐70]. Therefore, this study indicated that timely initiation of ANC and quality of ANC is vital to completion of maternal health care along the CoC.

The odds of completing CoC was found to be higher among women from communities with a high percentage of educated women. Low literacy levels in the community may be related to low health knowledge, while women from highly educated communities may acquire information from others regarding benefits of maternal health care, leading to increased odds of completing CoC. Illiterate women are economically unstable and may fail to receive adequate maternal health care during pregnancy [71].

Low proportions of women utilizing CoC in sSA were observed with the highest coverage in Southern Africa and almost similar low coverage in East, West and Central Africa. Both individual and community-level factors were associated with CoC completion. Factors associated with high CoC completion rates, include older age, having attended education, having mass media exposure, intended pregnancy, timely ANC initiation, female-headed households,, perceiving distance from health facilities not as big problem and residing in urban areas and from highly educated communities. These findings point to areas where care can be better tailored to improve the completion of CoC. Policymakers in sSA must consider both individual and community-level factors and undertake multi-sectorial approaches to address barriers at different levels. Thus, those living in rural areas, less educated, initiated ANC lately, perceiving distance from health facilities as big problem and from communities with low education need more attention to increase completion of CoC and improve maternal and newborn health.