Abstract
Malaria infection and deaths caused by the disease among children under five years in Africa are high. The disease contributed to 627,000 deaths, with children under the age of five years taking 80% of all malaria deaths in Sub-Saharan Africa. Tanzania is among the countries that are worst hit, alongside Mozambique, the Democratic Republic of the Congo, and Nigeria. In 2017, deaths for children under five were at 5% while incidents of under-5-year cases in Tanzania were at 37%. This study investigated whether the mother’s age, experience, education, and literacy levels have a relationship with mortality among infants under the age of five (5) years in Tanzania.
The results showed that the mean age of the respondents was 28.72 ± 9.64 years. Further, there was a statistically negative correlation between children under five mortality and (educational attainment and literacy); however, though negative, the correlation between under-5 mortality and experience was not significant. Further, findings revealed a significant positive association between under-5 mortality and women’s age. Lastly, the multiple linear regression model was found to be adequate and significant (R² = .637, F(4, 10018) = 97.217, p < .000).
Background information
World Health Organization statistics indicate that the disease is a significant burden in approximately 32 countries in Sub-Saharan Africa. This projection accounted for 95% of the 241 million malaria cases in 2020. Evidence indicates that the disease is caused and spread by the bite of infected female Anopheles mosquitoes. Furthermore, perennial flooding in many parts of African countries often hinders the fight against malaria.
This study chose Tanzania because of its peculiarity: it is one of the four countries in the African continent that accounts for more than half of all malaria deaths, alongside Nigeria, the Democratic Republic of Congo, and Mozambique. Studies show that the entire population of Tanzania’s mainland is at risk for the disease because about 93% of people in this region are considered to be living in Malaria transmission areas. The country experiences three malaria transmission seasons: stable perennial, stable malaria, and unstable seasonal.
Aim of the Research
Between 2017 and 2020, the incidence of the disease declined from 124 to 120 cases per 1,000 people considered to be at risk in Tanzania. However, deaths increased by 32% from 0.42 to 0.43 for every 1000 population at risk. During the same period, disease prevalence varied in different regions on the mainland. Children under 5 are mostly still cared for by their mothers, and so any infection affecting them is to a large extent attached to the mothers.
Data set and Year
The study will rely on the Demographic and Health Survey (DHS) for Tanzania. The data were collected on malaria indicators in the country in 2017. The data generated covered household information, including household recode, individual recode, children’s recode, household member recode, wealth index, malaria indicators, and other characteristics. Furthermore, data were generated for all regions in mainland Tanzania and Zanzibar.
Variables
In this study, there were two sets of variables, namely dependent (predictor or outcome) and independent (explanatory). Generally, the variables included mortality among children under five years, women’s age, educational attainment, literacy levels, and experience. In this case, the response variable was the mortality among children under 5 years in Tanzania. On the other hand, the explanatory variables included women’s age, education, literacy, and experience.
Methodology
This study utilized the DHS for Tanzania, a household survey conducted to gather data on malaria indicators among children under the age of five. The study fell into the category of standard DHS because it had a sample size of 10,018 and was usually carried out approximately every 5 years. It was rolled out in 2017, and for this study, the researcher incorporated datasets such as the age of the women, their education and literacy levels, and their experience. Furthermore, the survey research technique of conducting interviews with household members, using a questionnaire tool, was employed to gather data from the respondents.
Data Analysis Plan
The data analysis process employed both descriptive and inferential statistics. The characteristics of the data sets extracted from the Malaria Indicator Survey were summarized in terms of means, standard deviations, frequencies, and percentages. On the other hand, bivariate analysis was done to confirm the relationship between mortality among children under the age of 5 years and each of the independent variables (age, education, literacy levels, and experience).
The Pearson Product-Moment Correlation test was used. Additionally, a multiple linear regression analysis was conducted between the dependent variable and the three independent variables (women’s age, education, and experience) to determine the relative contribution of each predictor to the variation in the relationship. P-value and R-squared statistics were used to test the significance of the relationship and model adequacy.
Results
This survey used a sample size of 100,018 women aged between 15 and 49 years, with a mean age of the respondents of 28.72 ± 9.64 years. In terms of educational levels, 1827 (18.2%) had no education, 1459 (14.6%) had incomplete primary education, 4396 (43.9%) completed primary education, 1108 (11.1%) dropped out at the secondary level, 1100 (11%) completed high school education, while just about 128 (1.3%) attended post-secondary training. As far as their literacy was concerned, 6,673 (66.6%) were able to read a whole sentence, 653 (6.5%) could read only parts of a sentence, and 2,683 (26.8%) could not read at all. When asked about their experience in protecting their children from malaria, 8,801 (91.2%) confirmed having a good understanding of how to manage the disease, 671 (6.7%) did not have this experience, and about 546 (5.1%) were uncertain about their own knowledge.
A bivariate Pearson Product-Moment Correlation was run for under five child mortality and the four independent variables (age of women, educational attainment, literacy, and experience). There was a positive correlation between women’s age and the mortality of children under five, which was statistically significant (r = 0.070**, n = 10018, p = 0.001). Further, results showed a negative correlation between literacy and under-5 mortality which was significant at (r = -.156**, n = 10018, p =.000). Additionally, the findings revealed a negative correlation between the educational attainment of mothers and children under five deaths, which was statistically significant (r = -.184**, n = 10018, p =.000). Lastly, the correlation between women experience and mortality rate for children under-5 was negative, however, it was not statistically significant at (r = -.009, n = 10018, p =.379).
Multiple linear regression was used to quantify the relationship between the mortality response variable for children under five years and four predictor variables (women’s age, educational attainment, literacy, and experience). The overall regression was statistically significant (R2 =.637, F (4, 10018) = 97.217, p <.000). The fitted regression model would be (Under-5 mortality = 0.789 -.043 (experience) +.081 (women’s age) –.133 (educational attainment) -.069 (literacy).
References
“Malaria.” Who.int. Web.
Severe Malaria Observatory. Tanzania malaria facts. Web.
Sparkman D. Statistics: Descriptive vs. Inferential statistics. Study. Web.
Schober P, Christa B, & Lothar S. Correlation coefficients: Appropriate use and interpretation. Anesthesia & Analesia, 2018; 126 (5): 1763-1768. Web.
CFI Team. Multiple Linear Regression. CFI. Web.
Zach. The Complete Guide: How to Report Regression Results. Statology. 2021. Web.