Literature Review: Assessment of Malnutrition in Ethiopia Using Data Science Methods

Word Count: 2096

Introduction:

Making humans the center of human development as it is implied in the term requires that we focus on inherent individual rights because it is these underlying principles that serve as avenues to help meet the end of goal of human development- a satisfactory and enjoyable life. Amartya Sen captures these ideas perfectly in his book discussing development as freedom. In it, he explains how development is seen as the process of expanding real freedoms for people to enjoy, including economic freedom (ability to exchange goods in the market), social freedom (ability to pursue education and health care), and political freedom (broad political participation). Furthermore, he goes on to explain that when an institutional boundary deprives them of these basic, inherent freedoms, people no longer can make basic yet influential choices that both help them live enjoyable lives and create a flourishing society were everyone is satisfied. To that end, I believe that a major yet under-looked factor that is necessary for people to attain these inherent freedoms is access to necessities. Without food and water to survive, people can’t strive for that these freedoms, thereby preventing human development. Therefore, I have decided to focus my research on food malnutrition in Ethiopia, specifically from a data science lens. Subsequently, within my literature review, I will discuss the underlying issue related to Ethiopian malnutrition, assess the data science contributions made to this topic, and put forth a variety of implications that will help the audience better understand the state of the issue.

Analysis of Food Malnutrition and its relation to Ethiopia:

Under or Malnutrition is the health consequences associated with not taking in enough food energy and nutrition, resulting in hunger. It is not quality a person is inherently born with but rather a byproduct from the interactions of poor-quality diets, healthcare, environment, and poor behavior [4]. There 3 forms of malnutrition. They include stunting (someone who is too short for his/her age), wasting (acute malnutrition and rapid weight loss that makes it disproportionate to height), and underweight (someone who is low in weight for his/her age [4,7]. Globally, nearly 500 million adults are underweight, while nearly 200 million children under the age of 5 suffer from at one of the three forms of malnutrition. What’s worse is that 45% of deaths among young children are linked to some form of malnutrition [7]. Malnutrition is especially prevalent in developing countries. According to UNICEF and WHO, it contributes to the deaths of nearly 3 million children a year living, while continuously threatening the lives of millions each year. Being a developing country, malnutrition rates in Ethiopia are quite high, especially in those under the age of five; this is detrimental to the overall development of country, as survivors of malnutrition enter the world with social and mental disadvantages (suspectable to diseases, lack of cognitive development etc..). The most consistent predictors associated with child malnutrition in Ethiopia include low educational status of father and mother, poor household, low birth weight, and mothers ages (under 20) [4]. In the most consistent in child and adult malnutrition include unimproved sources of drinking water, absence of toilet facilities, infection, cancer, strenuous labor, lack of sufficient health care, and many others [4,8]. Consequently, instead of being a standalone issue, malnutrition is in fact interconnected to other problems plaguing our society. These interconnections also means that when malnutrition is combined with some other problem, the effects are synergistic. For example, when combined with cancer, the effects of both malnutrition and cancer together are greater than the effects of the individual effects summed up. Being such a complex issue, there is no one size fit all solution that policy makers or healthcare companies can directly apply to lessen its prevalence, but rather companies are to address the factors (stated above) that are responsible for malnutrition [4].

Data Science Contributions/Research methods

Many research contributions have been made to assess the prevalence of food malnutrition in Ethiopia. Based on further analysis of the research studies used in my annotated bibliography, it seems that across the board, researchers implement a similar three step process. They first survey their desired Ethiopian population to collect the correct test subjects that meet desired standard and exhibit measurable behavior. They collect data from such subjects. Second, researchers then analysis the data by comparing it to a given standard, using some sort of data comparison method. Finally, they discuss the results and draw conclusions based on them. To illustrate these steps, let me give an example. In 2016, the Central Statistical Agency (CSA) of Ethiopia investigated the magnitude of malnutrition in Ethiopian children under the age of five. Using a 2016 height and weight EDH (Ethiopian Demographic and Health) survey of 9494 child and mother pairings (51.3% male, 48.7% female, 21.4% under 12 months), the agency measured the nutritional status of Ethiopian children based on certain undernutrition indicators that abide by the WHO child growth standard. The measurable indicators used include stunting (child who is too short for his/her age), wasting (acute malnutrition and rapid weight loss that makes it disproportionate to height), and underweight (child who is low in weight for his/her age). WHO Stunting, wasting, and underweight standards serve as the dimensions of human development for this study, as the surveyed height and weight data is being compared to these standards. To make the comparative analysis, the agency used multilevel logistic linear regression adjusted for clusters and sampling weights to identify factors associated with the standards. The results showed that the prevalence of stunting was 38.3%, underweight, 23.3%, and wasting 10;1%. Seven out of fifteen (46.5%) suffer from at least on form of malnutrition while 3.1% suffered from all three kinds. Additionally, Sex of the child (male), children older than 24 months, recent experience of diarrhea, household wealth index (poorest) exhibit a higher risk of undernutrition. On the other hand, children born from overweight mothers and educated mother (primary, secondary or higher) exhibit a lower risk of undernutrition. With respect to the first step, the test subjects are the mother and child pairings while the data collected is 2016 survey data of height and weight. The original surveyed data consists of heights and weights of tens of thousands of children and mothers, but the researchers condense the data to the 9494 eligible pairs based upon a hierarchical classification data collection system. They stratify the data by looking at 11 distribution sites and then condense it to about 1200 enumerated areas. Following this, they then condense it to 18,008 households, and finally to the around 11,000 mother-child eligible pairings, all of which whose children seem to be malnourished. After extensive visitation, and interviewing of these pairings, the 9494 are chosen for analysis. This classification-based data method is popular among researchers looking to assess the prevalence health related issues within subjects, as it is tailored to ensure that the most appropriate subjects are tested. With respect to the second step, the standard to which the data is compared to are the 3 forms of malnutrition (seeing which of the 3 forms a given child), while the data analysis method to make such comparison is linear regression [4]. At the most basic level, linear regression is the process of graphing and scatter plotting two given behavioral data variables and measuring certain variable associated with these graphs (p-value, regression coefficient, constants etc…) in order to make thoughtful comparisons [5]. This basic form of data analysis is very common, as it is coding program friendly and easy to visualize. I will not provide an explanation on how this linear regression for this specific study works because I believe it is beyond the scope of this paper. With respect to the third step, the results are obviously the percentages, and the main conclusions are the characteristics that cause higher rates of undernutrition. Results are significant because they inform the companies and policy makers on what issues need to be addressed first. In this case, “underweighted” people must be addressed. On the other hand, the conclusions provide information that will drive their decision making. The conclusions for this study, which I believe are drawn based on an assessment of each pair (did a given pair have diarrhea etc..), and a comparison of that assessment to the likelihoods of having one of the forms of malnutrition, allow policy makers and companies to focus on improving diet habits (thereby fixing diarrhea), and building up the economy [4].

Though linear regression is the most popular data science method used among research studies; it is often not used alone. It is in fact coupled with other statistical methods, such as assessing standard deviation, medians, and confidence interval. In total, these methods provide a strong analysis of the prevalence of malnutrition within Ethiopian, providing policy makers with sufficient data to make appropriate change. This Data science analysis is not limited to assessing undernutrition in Ethiopian children, but has been expanded to assess adults, caner patients, and HIV patients [1,2,4,6]

Why I chose Ethiopia:

I have decided to focus my research on Ethiopia because it is one of the world’s poorest countries, making it a country in need of great public attention. In fact, 44% of its population live in poverty. Natural disasters, political conflict of some of the cause for poverty; the two main cause however has to do with its agricultural economy and difficulties of the world economy. Because agriculture is the primary source for Ethiopian economic growth, it is important for farmers to have sustainable agriculture. However, because smallholder farms form the largest poor group in the country, they not only have little access to the health care and education necessary for survival, but also lack basic farming infrastructure and tools that are needed to help grow better crops and subsequently foster economic growth. Droughts, overgrazing, and deforestation also degrade Ethiopian land, further inhibiting growth. Furthermore, Ethiopian poverty is also cause by rising global market prices on necessities, like food. Because of the increase of prices, households have limited resources and cannot purchase the food and fertilizer necessary for growth. Given, that country is plagued with a poor-agricultural economy and high prices, it is no wonder why malnutrition is becoming more and more prevalent in the country [3].

Gaps in research and Topics for Future Investigation:

Based on my scope research, most research studies use data science to assess the prevalence of undernutrition. However, not enough of it is used to help develop the viable solutions to malnutrition in. To that end, in the future, I hope to dive deeper into the solutions side of this issue and learn more about how data science is being used to help create solutions. Having synthesized multiple solutions, I hope to eventually determine which solution is best fitted for the given environment in Ethiopia.

Conclusion:

Today, issues of under or malnutrition plague Ethiopian society. This malnutrition is not a standalone issue but rather a byproduct of the other issues troubling the country, including poverty, rise in global prices, poor healthcare, high infectious rates, and an overall unsustainable agricultural economy. Today, data scientists use data classification methods and linear regression to assess the extent of malnutrition in Ethiopia, from which they draw conclusions and trends. Together, they provide policy makers and companies with sufficient data to make the necessary changes to help alleviate the problem. The data is there and solutions are up and rising so only time will tell when the malnutrition is completely eradicated from Ethiopia.

Works Cited [1] Alebel, A., Kibret, G.D., Petrucka, P. et al. Undernutrition among Ethiopian adults living with HIV: a meta-analysis. BMC Nutr 6, 10 (2020). https://doi.org/10.1186/s40795-020- 00334-x

[2 ]Gebremedhin, T. K., Cherie, A., Tolera, B. D., Atinafu, B. T., & Demelew, T. M. (2021). Prevalence and risk factors of malnutrition among adult cancer patients receiving chemotherapy treatment in cancer center, Ethiopia: cross-sectional study. Heliyon, 7(6), e07362. https://doi.org/10.1016/j.heliyon.2021.e07362

[3] Gomez, B. (2017, July 29). The Main Causes of Poverty in Ethiopia. borgenproject.org. https://borgenproject.org/main-causes-of-poverty-in-ethiopia/

[4] Kasaye HK, Bobo FT, Yilma MT, Woldie M (2019) Poor nutrition for under-five children from poor households in Ethiopia: Evidence from 2016 Demographic and Health Survey. PLoS ONE 14(12): e0225996. https://doi.org/10.1371/journal.pone.0225996

[5]Listen Data. (n.d.). 15 Types of Regression in Data Science. listendata.com. https://www.listendata.com/2018/03/regression-analysis.html#Linear-Regression

[6] Neima Endris, Henok Asefa, Lamessa Dube, “Prevalence of Malnutrition and Associated Factors among Children in Rural Ethiopia”, BioMed Research International, vol. 2017, Article ID 6587853, 6 pages, 2017. https://doi.org/10.1155/2017/6587853

[7] World Health Organization. (2021, June 9). Malnutrition. who.int. https://www.who.int/news-room/fact-sheets/detail/malnutrition

[8] US Aid. (2021, October 21). Nutrition. usaid.gov. https://www.usaid.gov/ethiopia/nutrition