Utilizing Data Science Methods to Improve Prediction of Maternal and Neonatal Care throughout Sub-Saharan Africa

by Suditi Shyamsunder

April 03, 2020

Introduction

The purpose of this research is to better understand the state of maternal and neonatal health care throughout the region of Sub-Saharan Africa. This research is extremely important because mothers and their children have the right to life regardless of where they are born and where they live. Those who live in Sub-Saharan Africa often have less access to the facilities and practices that they need to survive as compared to those in other regions of the world. Women who have to give birth in unsterile situations are less likely to survive and give birth to healthy children. Newborns who aren’t treated with the most up to date practices are less likely to survive the vulnerable time right after they are born. In fact, according to a study that looked at antenatal care practices on the survival rates of newborns in Sub-Saharan Africa, utilization of at least one antenatal care visit by a skilled provider during pregnancy reduced the risk of neonatal mortality by 39% in countries within this region [4]. There is so much that can be improved in terms of the care of newborns and their mothers, and it is important to continue research in this department. Amartya Sen discusses in his book ‘Development as Freedom’ that human development is all about removing unfreedoms. Helping mothers and newborns gain better access to healthcare throughout Sub-Saharan Africa removes unfreedom by providing these individuals the freedom of life. This freedom of life is the basic prerequisite of all other freedoms such as education and happiness. This work builds upon previous research by authors such as Armstrong, Bee, Bosomprah, Neal, and Tatem and aims to better utilize data science methodologies to understand maternal and neonatal care in the region of Sub-Saharan Africa.

Human Development Topic

Throughout the world, there are insufficient health care practices and facilities for many mothers and newborns. Women who are bringing new life in the world, should not have to do so on the side of the road because there is no health facility within walking distance, and no one should have to seriously wonder whether their newborn will stay alive even another few weeks. If this pain is preventable, it ought to be prevented. The first week of life is a particularly vulnerable time and 73% of neonatal deaths occur during this period. In developing nations in Sub-Saharan Africa, the quality of newborn and maternal care needs to be improved and standardized [2].

There is a large inequality between the quality of care for mothers and newborns in Sub-Saharan Africa as compared to those in more developed nations such as the United States. There is even a large amount of inequity in different regions of the same country. Under the status quo, the lack of sufficient care leads to loss of the lives that were just beginning. Pregnant women going into labor often have unrealistically far distances to travel if they wish to give birth in a proper facility and even those who make it to a facility are faced with unideal care practices. By improving neonatal care practices, millions of lives can be saved throughout the developing regions of Africa. Right now, newborn care practices are not up to par because resources are not being distributed where they need to be. More health care facilities are required in rural parts of Africa and knowledge of newer practices that have been proven to save more lives need to be spread to health care providers throughout the continent. Every year, 2.9 million neonates die and saving even some of these newborns would provide freedom of life and help achieve the goals of human development [2].

Amartya Sen’s definition of human development holds freedom at its core. His principle revolves around the idea that the most important part of development is how it expands people’s freedoms in an expansive sense. Bettering the facilities for maternal and newborn care definitely improves the freedom of life because surviving childbirth is far more likely with the proper sterility and doctors. This topic aligns best with the dimension of social opportunities. In his book he includes both education and health facilities as a subset of social opportunities because without these relatively basic necessities people do not have the capability to get a job or work to lift themselves into a more desirable situation. Maternal and newborn health care also relates well to the sustainable development goal 3, which is to ensure healthy lives and promote well-being for all at all ages. To me this goal strives to provide the best healthcare to those who need it including mothers and their babies that may not even yet be born.

Human Development Process

Healthcare in sub-saharan Africa is a very complex system dependent on many different factors. Unfortunately, a study that analyzed Gallup World Poll data found that in comparison to other regions of the world, sub-Saharan Africa has the lowest ratings for well-being and the lowest satisfaction with health care [5]. Lack of proper health care is associated with many factors such as less access to education and larger rates of poverty. Overall people throughout the region have access to much lower quality of life and healthcare as compared to other regions of the world. This inequality exists in income, poverty, and healthcare. In terms of maternal and neonatal care specifically, health opportunities for women in Sub-Saharan Africa are scarce and not up to par. In fact, the region accounts for about two-thirds of all maternal deaths in the world with a maternal mortality ratio (MMR) of 546 maternal deaths per 100 000 live births [9]. While maternal care is certainly influenced by broader factors such as education and income, some less obvious factors were also researched in one study that found a relationship between the maternal mortality ratio, sanitation, economic factors, prenatal care coverage, births assisted by skilled health personnel, access to an improved water source, adult literacy rate, primary female enrollment rate, education index, the Gross National Income per capita and the per-capita government expenditure on health. The high number of factors demonstrate the inherent complexity of maternal healthcare in Sub-Saharan Africa [7].

What makes healthcare even more complicated in the Sub-Saharan region is the variations and inequality that exist across the area. Individuals living in certain regions within Sub-Saharan Africa have much better access to care than those in less fortunate areas. This inequality extends specifically to maternal and neonatal care within Sub-Saharan Africa. Within Tanzania, in the region of Mwanza, 27% of women had reported birth complications, but in Dar es Salaam, 80% do [1]. This shows how even within the same country, there can be a large inequity in the quality of care available. In addition, this inequality is dependent on many factors, For example, results of one study show that births among women in the richest wealth quintile are 68% more likely to occur in health facilities than births among women in the lowest wealth quintile. Women with at least primary education are also twice as likely to give birth in facilities than women with no formal education [6].

In his presentation entitled ‘Development and Complexity’, Owen Barder discusses the idea of human development processes acting as complex adaptive systems. This is certainly the case for maternal and neonatal care in Sub-Saharan Africa. Much of the complexity originates from the fact that healthcare access is intertwined with education and income levels. The system is also definitely adaptive. In order for maternal health care to improve, the entire healthcare system must as well. And, in order for the healthcare system to improve, education must be wildly available for all people regardless of gender, social status, or other factors. Income inequality must also decrease in order to better the complex adaptive system that is healthcare in Sub-Saharan Africa.

Geospatial Data Science Methods

Throughout the research done on maternal and neonatal care in Sub-Saharan Africa, many different datasets were analyzed and utilized to draw conclusions. Much of the data was in the form of survey data. In addition, many different data science methodologies were used in order to interpret the data. Maps were drawn that show distributions and techniques such as Bayesian hierarchical regression models were utilized.

One study researched the quality of care provided by different facilities in Ghana for women and newborns. The facilities referenced in this article were being rated based on different essential functions regarding Emergency Obstetric and Newborn Care (EmONC). The results of this study uncovered that a mere 21% of births in facilities in Ghana occured in those that were fully functioning. The data for this study was collected from the Ghana EmONC survey, which was a “nationwide cross-sectional facility-based survey” that was conducted in 2010. The data analysis included creating maps that showed differences in the quality of health care facilities across the country of Ghana [3].

In order to get a greater understanding of the quality of maternal and newborn health in developing countries around the world, it is important to have a baseline understanding of the populations that exist in the studies area. One article focused on mapping the distributions of women of childbearing age, pregnancies, and birth. This is a particularly useful endeavor because discovering methods for safer births and healthier newborns could be improved by more accurate estimations of these populations. The study utilized various WorldPop data as well as UN statistics on age specific fertility rates, live births, stillbirths, and abortions to analyze and map the distributions in 4 main countries: Afghanistan, Bangladesh, Ethiopia, and Tanzania [10].

Another study utilized Demographic and Health Surveys (DHS) in order to achieve a similar goal. The researchers analyzed the data from first births before the age of 20 but also disaggregated it into three groups: under 16, 16-17, and 18-19 years of age. Using this data combined with GPS-located cluster data and adaptive bandwidth kernel density estimates, they were able to create descriptive choropleths, and prevalence maps. They also used a Bayesian hierarchical regression modeling approach to map adolescent first birth at a district level with estimates of uncertainty. A Bayesian hierarchical regression model is a statistical model that allows data scientists to make conclusions about their data using Bayes Theorem. In this case, it was used to discover the distribution of first births to young mothers in these countries [8].

Discussion

Much research has been done on the status of the healthcare system in Sub-Saharan Africa but there is also more to be analyzed and understood. Specifically, the bulk of the research focuses on two main areas. The first of these is research on the quality of facilities across the region. The second is the distribution of individuals in need of maternal health care across Sub-Saharan Africa.

While it is essential to understand the distribution of potential mothers as well as health care facilities, I feel that a combination of this research could be effective in understanding how to better the health care system. It is more useful to understand where the highest concentrations of mothers are in order to determine where functioning health care facilities are most needed. It is important to see whether or not the distribution of the need, in this case pregnant mothers, is aligned with the distribution of supply, in this case facilities.

It would be beneficial to do research on the distance that mothers need to travel in order to get access to a facility to give birth in. This research would allow us to see the distribution and any inequality that exists in different parts of Sub-Saharan Africa. It is possible that the facilities with the best care are concentrated in the more wealthy areas, and it is important to be able to see this distribution. With this information, policy makers may be more likely to know where money to build new facilities should be spent. In this way, the resources that we have could be used where they are most needed. While much research has been done on understanding the quality of neonatal and maternal healthcare in Sub-Saharan Africa, further study would be beneficial and could change the lives of many.

Works Cited

[1] Armstrong, Corinne E., et al. “Subnational Variation for Care at Birth in Tanzania: Is This Explained by Place, People, Money or Drugs?” BMC Public Health, vol. 16, no. S2, 2016.

[2] Bee, Margaret, et al. “Neonatal Care Practices in Sub-Saharan Africa: a Systematic Review of Quantitative and Qualitative Data.” Journal of Health, Population and Nutrition, vol. 37, no. 1, 2018.

[3] Bosomprah, Samuel, et al. “Spatial Distribution of Emergency Obstetric and Newborn Care Services in Ghana: Using the Evidence to Plan Interventions.” International Journal of Gynecology & Obstetrics, vol. 132, no. 1, 2015.

[4] Chojenta, Catherine, et al. “The Impact of Antenatal Care on Neonatal Mortality in Sub-Saharan Africa: A Systematic Review and Meta-Analysis.” PLOS ONE, Public Library of Science.

[5] Deaton, Angus S, and Robert Tortora. “People in Sub-Saharan Africa Rate Their Health and Health Care among the Lowest in the World.” Health Affairs (Project Hope), U.S. National Library of Medicine, Mar. 2015.

[6] Doctor, Henry V, et al. “Health Facility Delivery in Sub-Saharan Africa: Successes, Challenges, and Implications for the 2030 Development Agenda.” BMC Public Health, BioMed Central, 19 June 2018.

[7] Lerberghe, W. Van, et al. “Factors Associated with Maternal Mortality in Sub-Saharan Africa: an Ecological Study.” BMC Public Health, BioMed Central, 1 Jan. 1970.

[8] Neal, Sarah, et al. “Mapping Adolescent First Births within Three East African Countries Using Data from Demographic and Health Surveys: Exploring Geospatial Methods to Inform Policy.” Reproductive Health, vol. 13, no. 1, 23 Aug. 2016.

[9] Pons-Duran, Clara, et al. “Inequalities in Sub-Saharan African Women’s and Girls’ Health Opportunities and Outcomes: Evidence from the Demographic and Health Surveys.” Journal of Global Health, Edinburgh University Global Health Society, June 2019.

[10] Tatem, Andrew J, et al. “Mapping for Maternal and Newborn Health: the Distributions of Women of Childbearing Age, Pregnancies and Births.” International Journal of Health Geographics, vol. 13, no. 1, 4 Jan. 2014.