Women’s literacy and children’s mortality rate among Southeast Asia countries 1991-2020: a cross-sectional study

Introduction

The global under-five mortality rate (UFMR) declined by 59% from 93 deaths per 1000 live births in 1990 to 39 in 2018. However, the burden of under-five deaths remains unevenly distributed. About 74% of under-five deaths occurred in two regions in 2018, Africa (52%) and South-East Asia (22%). The highest under-five mortality rate remains in the African Region (76 per 1000 live births), around 8 times higher than that in the European Region (9 per 1000 live births).¹ This number needs to be improved as the United Nations’ Sustainable Development Goals (SDGs) target for 2030 is to reduce preventable under-five mortality to at least as low as 25 per 1000 live births. However, achieving this goal becomes more challenging as most under-five mortality occurs in low (LICs) and lower-middle-income countries (LMICs), especially countries within Africa and South Asia,^1,2 where children’s health can be affected by several factors such as healthcare service quality, family wealth status, and social status, including mother’s education level.³

Many kinds of research had been conducted to identify the association between a mothers’ education level and their children’s health status. The evidence has shown that the mother’s education is an important determinant of the health of children.⁴ It’s stated that even after accounting for household income, the number of siblings, health environments, and other socioeconomic variables, the mother’s literacy is still a major factor that influences children’s health status.⁴ A study conducted in the Kashmir valley has shown an inverse relationship between women’s literacy rate (WL) and children’s mortality rate (CMR). This study found that women’s literacy has an immense contribution in declining the infant mortality rate (IMR) and maternal mortality rate (MMR) and thus can help in improving the health status of both women and children.⁵ In sub-Saharan Africa, a study found that the decline in mortality rates of children under five years was much higher among the children born to mothers who have never received formal education.⁶ In Indonesia, child vaccination rates were increased from 19% when mothers have no education, compared to 68% when mothers have at least secondary school education.⁷ A study in India also showed that the female literacy rate is a good predictor of infant mortality rate in India. The study’s results showed that infant mortality rate was inversely related to women’s literacy rate, while men’s literacy was not.⁸

Proper education is the first step in empowering women and young girls and providing women with the best possible chance for a prosperous and healthy life.⁹ For women who go on to have families, education can also aid in household and family management.¹⁰ Lastly, educating girls saves lives and builds stronger families, communities, and economies. In addition, an educated female population increases a country’s productivity and promotes economic growth.¹¹ According to an International Labor Organization report, educating girls has proven to be one of the most important ways of breaking poverty cycles and is likely to have significant impacts on access to formal jobs in the longer term.¹¹ Moreover, it’s reported that some countries lose more than $1 billion a year by failing to educate girls to the same level as boys. Therefore, it’s believed that educating women will increase their opportunity to get a better job and receive a higher wage. Furthermore, this also may address gender imbalances especially in the labor force.¹¹ Since literate women are considered capable of understanding written information and using it to improve the health, nutrition, and education of household members, and after looking at the importance and benefits of women’s literacy in various sectors, especially health, we were interested in conducting this study.¹²

Methods

Results

Figure 1 showed that CMR trends in 11 Southeast Asian countries had varied over the past 30 years. However, it was generally seen that the CMR in 2020 was lower than in 1991. Then, we developed Table 1 to describe the average of WL, and CMR (IMR, NMR, and UFMR) among 11 Southeast Asian countries from 1991 to 2020. Meanwhile, Tables 2 and 3 were made to show the results of data normality test and Kruskal-Wallis test in this study. Table 1 described that Philippines (89.3), Thailand (89.7), Singapore (89.3), and Brunei Darussalam (89.3) were the countries with the highest average WL in the region, while Cambodia (64.5), Lao (59.6), and Timor Leste (40.8) were the lowest. Then, Singapore (2.9, 1.5, 3.6), Malaysia (8.5, 4.8, 9.9), and Brunei Darussalam (8.7, 5.2, 10.8) were the three countries with the lowest average of CMR (IMR, NMR, and UFMR) among countries in Southeast Asia, while Timor Leste (70.5, 32.6, 88.6), Lao (64.5, 33.4, 88.6), and Myanmar (56.9, 33.3, 77.1) were consistently the highest in the last 3 decades.

Figure 1. The trend of CMR in 11 Southeast Asian Countries from 1991 to 2020.

Next, since from Table 2 it’s known that not all variables had normal distribution, therefore, we conducted Kruskal-Wallis test instead of ANOVA and the result of this test was provided in Table 3. After knowing that there was significant difference between variables, then we run post-hoc test to assess the specific differences between variables. The full results from this test were provided in Tables 3.1 to 3.4. Moreover, from correlation test it’s found that there was significant negative correlation between WL and CMR in all countries except Brunei Darussalam. In Brunei, WL only significantly correlated with UFMR (-0.520, p=0.003), while other countries showed a significant correlation between WL and IMR, NMR, and UFMR with varying degrees of correlation. From Table 4, it can be identified that the highest correlation between CL and CMR were found in Cambodia (IMR=-0.962, NMR=-0.965, UFMR=-0.95, p<0.001), Indonesia (IMR=-0.906, NMR=-0.91, UFMR=-0.895, p<0.001), and Vietnam (IMR=-0.861, NMR=-0.871, UFMR=-0.843, p<0.001).

Then, we conducted linear regression analysis to indicate how IMR, NMR, and UFMR in the Southeast Asian region were explained by WL of those 11 countries. The result of this test was available in Table 5. From the table, it’s known that in Southeast Asia, the correlation between WL and IMR is the highest (R=0.805), where around 64.8% of IMR can be explained by WL (R²=0.648, F=602.8, p<0.001). Meanwhile, WL was only able to explain 54.4% and 60.9% of NMR (R²=0.544, F=391.7, p<0.001) and UFMR (R²=0.609, F=511.7, p<0.001), respectively.

Discussion

In 2018, most of under-five deaths occurred in Africa (52%) and South-East Asia (22%).¹ Our study results showed that although there has been a decline in UFMR by more than 50% in the last 30 years as on a global scale, Figure 1 shows that the downward trend of UFMR has slowed since the early 2000s. Furthermore, it’s reported that across all regions, the annual rate of reduction from 1990 to 2020 was larger for children aged 1–59 months than for newborns.¹⁷ Figure 1 also showed that similar to global scale, in Southeast Asian countries, the NMR and IMR have a smaller downward trend than UFMR and even relatively unchanged in some countries in the region. Moreover, during the COVID-19 pandemic, the United Nations Inter-Agency Group for Child Mortality Estimation (UN IGME) found no significant evidence of excess mortality among children, adolescents and youth in 2020.¹⁸ Although many limitations were raised regarding the report, it is difficult to assume that the slowing down of the CFR trend since the early 2000s is related to the Severe Acute Respiratory Syndrome (SARS) outbreak in that period.

Next, in this study we found that Timor Leste, Lao, and Myanmar have the highest average of CMR across country in the region, while Singapore, Brunei Darussalam, and Malaysia had the lowest rates of children mortality compared to other countries in this region. If referring to the World Bank’s criteria, the 3 countries with the highest CMR in this study are low-income countries, while the 3 countries with the lowest CMR are high-income countries. Therefore, this study results is in line with the previous researches which showed that the highest burden of under-five mortality is reported to be mostly found in lower-income countries,¹⁹ while in contrast, the countries with the higher-income have the lowest.²⁰

Through Kruskal-Wallis test, our study found differences in mean WL, IMR, NMR, and UFMR across 11 countries in the Southeast Asian region in the last 3 decades. Then, based on our post hoc test results, it was found that higher-income countries are often different from lower-income countries. For example, no significant differences were found between the average of WL in Timor Leste and Laos, Myanmar and Cambodia, and even Vietnam and the Philippines. However, significant differences were found in the average of WL between Timor Leste and Singapore, Malaysia, Brunei Darussalam, and even Thailand. The same trend was also found in other variables (IMR, NMR, and UFMR) in this study. This finding is coherent to previous studies that have shown a strong relationship between a country’s income level and children’s mortality rates.²⁰

Our study wanted to analyze the relationship and impact of WL on CMR in Southeast Asian countries in the last 30 years. Then our linear regression results showed that the correlation between the WL and the IMR was the highest, followed by WL and UFMR, and WL and NMR. However, while the correlation between WL and IMR reached more than 80%, but it’s only around 65% of IMR can be explained by WL. The correlation between WL and NMR and UFMR is even lower and can only explain 54% and 61% of each variable.

Women who have literacy are believed to be able to use their knowledge and understanding to improve the welfare of their families.¹² In addition, the previous study had been proven that better education of women might reduce mortality among children.²¹ However, our study showed that WL is not the best predictor of CMR. The linear regression result in this study showed that correlations (R) between WL and IHR, NMR, and UFMR were 81%, 74%, 78%, meanwhile the R² values were 65%, 54%, and 61% respectively. This finding showed how WL has significant correlation and impact on CMR.

Based on this study results, we assume that to achieve the SDG target in 2030, it is necessary to accelerate the reduction of CMR which can be done through strengthening in various sectors, including governance, the economy, health care, and education, as well as by strengthening gender equality. Accelerated improvement in various sectors is urgently needed, considering that the latest SDG report mentions that many SDG indicators have experienced setbacks due to the impact of the COVID-19 pandemic.²² In health sector, a recent report mentioned that while there has been major progress in the coverage of skilled health workers since 2000, more progress is still needed in high-burden areas. Moreover, although more and more deliveries were assisted by expert health workers, the coverage is not evenly distributed with significant disparities between regions.²³ In governance sector, freedom and democracy are deemed necessary to reduce the CMR figure. A previous study showed how the level of democracy was negatively associated with under-5 mortality, and that that negative association is greater in the presence of media freedom.²⁴ Furthermore, as the governance of a country may have widespread effects on the health of its population.²⁵ A previous study had shown that the better the governance, the lower CMR. In another study, it’s mentioned that the quality of governance is even more critical in determining a good outcome for both mother and child.²⁶

Conclusion

Our results show that there is a variation in CMR trends among countries in Southeast Asia in the last 30 years. Then, we also found that there was a significant relationship between WL and CMR. However, based on the results of our analysis, we conclude that WL is not the only factor that can explain the level of CMR in a country, instead other factors such as country’s income level, governance, and health system also need to be considered. Further research is needed to prove this assumption.

Data availability