Socioeconomic, urban‐rural and sex‐based inequality in infant mortality rate: evidence from 2013 Yemen demographic and health survey

Archives of Public Health, Volume 79, No. 1, Year 2021

Interpretation

Background: The occurrence of Infant Mortality Rate (IMR) varied globally with most of the cases coming from developing countries including Yemen. The disparity in IMR in Yemen however, has not been well dealt and therefore we examined the IMR inequality based on the most reliable methodology in order to generate evidence-based information for some program initiatives in Yemen. Methods: Based on the World Health Organization (WHO) Health Equity Assessment Toolkit (HEAT) software, we analyzed the inequality across the different inequality dimensions in Yemen. The toolkit analyzes data stored in the WHO health equity monitor database. Simple and complex, and absolute and relative measures of inequality were calculated for the four dimensions of inequality (subpopulations) which included wealth, education, sex and residence. We computed a 95 % CI to assess statistical significance. Results: The analysis included 31, 743 infants. Absolute and relative wealth-driven, education, urban-rural and sex-based inequalities were found in IMR. Higher concentration of IMR was observed among infants from the poorest/poor households (ACI=-4.68, 95 % CI; -6.57, -2.79, R = 1.61, 95 % CI; 1.18, 2.03), rural residents (D = 15.07, 95 % CI; 8.04, 22.09, PAF=-23.57, 95 % CI; -25.47, -21.68), mothers who had no formal education (ACI=-2.16, 95 % CI; -3.79, -0.54) and had male infants (PAF= -3.66, 95 % CI; -4.86, -2.45). Conclusions: Higher concentration of IMR was observed among male infants from disadvantaged subpopulations such as poorest/poor, uneducated and rural residents. To eliminate the observed inequalities, interventions are needed to target the poorest/poor households, rural residents, mothers with no formal education and male infants. The study was based on large data set stored in the WHO health equity monitor database for Yemen republic. Yemen is among the poorest country in the Middle East and North Africa [19], where the process of political transition has risen into a complete civil war since early 2015 [19, 20]. The war consequently led to internal disruption of services of all sectors among which healthcare services, displacement, causalities, and destruction of infrastructures such as main roads, bridges and airports [19, 20] are among the major ones. In the process, more than 12 million people experience food insecurity, 20.4 million of them deprived of safe water and adequate sanitation with more than 1.8 million children lost their school access. In addition, about 80 % or 21.1 of the people requiring humanitarian aids and protection [19, 20] particularly children since most are grappling with mass outbreaks of preventable diseases such as cholera, diphtheria, measles and dengue fever. Based on the projection evidences of previous data, IMR as of 2020 is estimated to be 43 per 1000 live birth [21, 22]. We used a nationally representative data extracted from 2013 Yemen Demographic and Health Survey (YDHS), through the offline version of the WHO 2019 updated Health Equity Assessment Toolkit (HEAT) software (Available for free at https://www.who.int/data/gho/health-equity/heat-built-in-database–edition.) [23]. Although the detail discussion of the software is available elsewhere [23–25], the software is summarized as a toolkit which enables to scrutinize and analyze data on health inequalities within and between countries [26]. The software is tremendously valuable to explore the health disparity situation in a more systematic detail and comprises of the WHO Health Equity Monitor (HEM) database [26]. The database stores data coming from Demographic and Health Survey (DHS) and Multiple Indicator Cluster Survey (MICS) conducted in many low-or-middle income countries including Yemen. Currently, the database provides details of inequality assessment for more than 30 Reproductive, Maternal, Newborn and Child health indicators including IMR [24, 25]. The YDHS done in all the 20 governorates of the Republic including the capital city Sana’a with the aim to provide planners and decision makers with information on various health topics that includes IMR though in the survey, largest governorates were under sampled while the smallest ones were oversampled to produce sufficient samples for each governorate and the capital city. The 2013 YDHS sample was selected using a stratified two-stage cluster design consisting of 800 clusters, with 213 in urban areas and 587 in rural areas [27]. The Birth Recode (BR file) was used for analysis of IMR [27]. IMR is the variable of interest that we measured inequality of and refers to the probability (expressed as a rate per 1,000 live births) of a child exposed in a specific period dying before reaching their first birthday or the number of child death before celebrating first birth day per 1000 live births [28, 29]. The survey collected full birth histories for women aged 15–49 and years and children (including date of birth [28, 29]. A full birth history is a complete list of all children the woman has ever given birth which included date of birth, sex, survival status, age (if alive), and age at death (if died) [28, 29]. Birth histories captured all live births, including children who later died, but omitted stillbirths, miscarriages or abortions. Birth histories are collected in chronological order from first to last. Children born five years preceding the survey were included in the analysis [28, 29]. Inequality in IMR was measured in relation with four dimensions of inequality (sub-populations) that included economic status, education status, place of residence and sex (infant sex). Economic status was measured using a wealth index. The DHS constructs wealth index based on different household materials and durable assets collected during the survey. DHS uses Principal Component Analysis (PCA) to create the index and detail description on the creation of the wealth index is published elsewhere [30]. The wealth index was devided into five quintiles: poorest, poor, middle, rich and richest. Education refers to the highest level of schooling attained by the woman and is categorized as no education, primary school and secondary school and above, place of residents as rural vs. urban, child sex as male vs. female and subnational region included 20 region and one city (Ibb, Abyan, Sanaa City, Al-Baidha, Taiz, Al-Jawf, Hajjah, Al-Hodiedah, Hadramout, Dhamar, Shabwah, Sadah, Sana’a, Aden, Lahj, Mareb, Al-Mhweit, Al-Mhrah, Amran, Aldhalae and Reimah). We used the 2019 update WHO-HEAT software for data analysis and performed in steps. First, we disaggregated the IMR disparities by the above-mentioned five equity stratifiers (economic status, education status, infant sex, place of residence and subnational region). Then, we calculated summary measures; Difference (D), Ratio (R), Population Attributable Fraction (PAF) and Absolute Concentration Index (ACI) for each equity stratifiers. However, subnational region inequality in IMR was not calculated for all four summary measures either due to missing or unavailable data in two regions (Mareb and Al-Mhrah). Since ACI is applied only for ordered equity stratifiers such as economic and education status, it was not calculated for place of residence and sex. Details about calculation and interpretations of summary measures are described in the WHO health inequality handbook [18] and technical notes of HEAT software [23]. For some clarity on ACI, it is a complex, weighted measure of inequality that shows the health gradient such as IMR across multiple subgroups with a natural ordering, on an absolute scale and indicates the extent to which IMR was concentrated among the disadvantaged or the advantaged group. If there is no inequality in IMR, ACI takes the value zero. Positive values indicated a concentration of IMR among the advantaged subpopulation (richest and secondary school and above for economic and education status inequality dimensions, respectively), while negative values indicate a concentration of IMR among the disadvantaged subpopulation (poorest and no formal education for economic and education status inequality dimensions, respectively). ACI is usually negative for unfavorable health indicators such as IMR. The larger the absolute value of ACI, the higher the level of inequality of IMR. Difference (D) is a simple, un-weighted measure of inequality that shows the absolute difference between two subgroups. If there was no inequality in IMR, D takes the value of zero. Greater (in absolute values) indicates higher levels of inequality in IMR. A positive value indicated a higher concentration of IMR among infants in the disadvantaged subpopulations (poorest, no formal education, rural residents and male for economic status, education status, place of residence and sex (infant sex) inequality dimensions respectively) and a negative value indicated that higher concentration of IMR among infants in the advantaged subpopulations (richest, secondary school and above, urban residents and female for economic status, education status, place of residence and sex (infant sex) inequality dimensions, respectively). PAF is a complex, weighted measure of inequality that shows the potential for improvement in the national level of IMR, in relative terms, that could be reduced if all subgroups had the same level of IMR as a reference subgroup. PAF was calculated by dividing the population attributable risk (PAR) by the national average µ of IMR and multiplying the fraction by 100. PAF takes negative values for adverse health outcome indicators like IMR. The larger the absolute value of PAF, the larger the degree of inequality. PAF is zero if no further reduction in IMR could be achieved, i.e. if all subgroups have reached the same level of IMR as the reference subgroup. The reference groups are subpopulation with lowest estimate of IMR; richest, secondary school and above, urban residents and female for economic status, education status, place of residence and sex (infant sex) inequality dimensions, respectively. Ratio (R) is a simple, un-weighted measure of inequality that shows the relative inequality between two subgroups. If there is no inequality, R takes the value one. It takes only positive values and the further the value of R from 1, the higher the level of inequality. The demographic health surveys are available publicly and ethics approvals were completed by institutions that commissioned, funded, and managed the surveys. DHS surveys are approved by Inner City Fund (ICF) International and an in-country Institutional Review Board (IRB) to ensure protocols are in compliance with the U.S. Department of Health and Human Services regulations for the protection of human subjects. It is however worth noting that, in this study, we used publicly available DHS data which did not require further ethical clearance.

AI Digest

Study Justification:

The study aimed to examine the inequality in Infant Mortality Rate (IMR) in Yemen, a developing country with high IMR. The study used reliable methodology to generate evidence-based information for program initiatives in Yemen.

Highlights:

– The study analyzed data from the World Health Organization (WHO) Health Equity Assessment Toolkit (HEAT) software.
– Inequality in IMR was examined across different dimensions, including wealth, education, sex, and residence.
– The study found absolute and relative inequalities in IMR based on these dimensions.
– Higher concentration of IMR was observed among infants from the poorest/poor households, rural residents, mothers with no formal education, and male infants.
– The study highlights the need for interventions to target these disadvantaged subpopulations in order to eliminate the observed inequalities.

Recommendations:

– Target interventions towards the poorest/poor households, rural residents, mothers with no formal education, and male infants to address the observed inequalities in IMR.
– Implement programs to improve access to healthcare services, education, and economic opportunities for these disadvantaged subpopulations.

Key Role Players:

– Government agencies responsible for healthcare, education, and social welfare.
– Non-governmental organizations (NGOs) working in the field of maternal and child health.
– Community leaders and local organizations.
– International organizations providing humanitarian aid and support.

Cost Items for Planning Recommendations:

– Healthcare infrastructure development and improvement.
– Training and capacity building for healthcare professionals.
– Education programs and initiatives.
– Economic empowerment programs for disadvantaged households.
– Outreach and awareness campaigns.
– Monitoring and evaluation systems to track progress and impact of interventions.
Note: The actual cost of implementing these recommendations will depend on the specific context and resources available in Yemen.

Strength of Evidence

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong because it is based on a large dataset from the WHO health equity monitor database. The study used the World Health Organization (WHO) Health Equity Assessment Toolkit (HEAT) software to analyze the inequality across different dimensions in Yemen. The analysis included a significant number of infants (31,743) and found absolute and relative wealth-driven, education, urban-rural, and sex-based inequalities in infant mortality rate (IMR). The study provides specific measures of inequality (Difference, Ratio, Population Attributable Fraction, Absolute Concentration Index) and includes 95% confidence intervals to assess statistical significance. The conclusions highlight the need for interventions to target the poorest/poor households, rural residents, mothers with no formal education, and male infants to eliminate the observed inequalities. To improve the evidence, it would be beneficial to provide more details on the methodology used, such as the specific variables and statistical tests employed. Additionally, including information on potential limitations and biases in the study would enhance the overall strength of the evidence.

Abstract

The study was based on large data set stored in the WHO health equity monitor database for Yemen republic. Yemen is among the poorest country in the Middle East and North Africa [19], where the process of political transition has risen into a complete civil war since early 2015 [19, 20]. The war consequently led to internal disruption of services of all sectors among which healthcare services, displacement, causalities, and destruction of infrastructures such as main roads, bridges and airports [19, 20] are among the major ones. In the process, more than 12 million people experience food insecurity, 20.4 million of them deprived of safe water and adequate sanitation with more than 1.8 million children lost their school access. In addition, about 80 % or 21.1 of the people requiring humanitarian aids and protection [19, 20] particularly children since most are grappling with mass outbreaks of preventable diseases such as cholera, diphtheria, measles and dengue fever. Based on the projection evidences of previous data, IMR as of 2020 is estimated to be 43 per 1000 live birth [21, 22]. We used a nationally representative data extracted from 2013 Yemen Demographic and Health Survey (YDHS), through the offline version of the WHO 2019 updated Health Equity Assessment Toolkit (HEAT) software (Available for free at https://www.who.int/data/gho/health-equity/heat-built-in-database–edition.) [23]. Although the detail discussion of the software is available elsewhere [23–25], the software is summarized as a toolkit which enables to scrutinize and analyze data on health inequalities within and between countries [26]. The software is tremendously valuable to explore the health disparity situation in a more systematic detail and comprises of the WHO Health Equity Monitor (HEM) database [26]. The database stores data coming from Demographic and Health Survey (DHS) and Multiple Indicator Cluster Survey (MICS) conducted in many low-or-middle income countries including Yemen. Currently, the database provides details of inequality assessment for more than 30 Reproductive, Maternal, Newborn and Child health indicators including IMR [24, 25]. The YDHS done in all the 20 governorates of the Republic including the capital city Sana’a with the aim to provide planners and decision makers with information on various health topics that includes IMR though in the survey, largest governorates were under sampled while the smallest ones were oversampled to produce sufficient samples for each governorate and the capital city. The 2013 YDHS sample was selected using a stratified two-stage cluster design consisting of 800 clusters, with 213 in urban areas and 587 in rural areas [27]. The Birth Recode (BR file) was used for analysis of IMR [27]. IMR is the variable of interest that we measured inequality of and refers to the probability (expressed as a rate per 1,000 live births) of a child exposed in a specific period dying before reaching their first birthday or the number of child death before celebrating first birth day per 1000 live births [28, 29]. The survey collected full birth histories for women aged 15–49 and years and children (including date of birth [28, 29]. A full birth history is a complete list of all children the woman has ever given birth which included date of birth, sex, survival status, age (if alive), and age at death (if died) [28, 29]. Birth histories captured all live births, including children who later died, but omitted stillbirths, miscarriages or abortions. Birth histories are collected in chronological order from first to last. Children born five years preceding the survey were included in the analysis [28, 29]. Inequality in IMR was measured in relation with four dimensions of inequality (sub-populations) that included economic status, education status, place of residence and sex (infant sex). Economic status was measured using a wealth index. The DHS constructs wealth index based on different household materials and durable assets collected during the survey. DHS uses Principal Component Analysis (PCA) to create the index and detail description on the creation of the wealth index is published elsewhere [30]. The wealth index was devided into five quintiles: poorest, poor, middle, rich and richest. Education refers to the highest level of schooling attained by the woman and is categorized as no education, primary school and secondary school and above, place of residents as rural vs. urban, child sex as male vs. female and subnational region included 20 region and one city (Ibb, Abyan, Sanaa City, Al-Baidha, Taiz, Al-Jawf, Hajjah, Al-Hodiedah, Hadramout, Dhamar, Shabwah, Sadah, Sana’a, Aden, Lahj, Mareb, Al-Mhweit, Al-Mhrah, Amran, Aldhalae and Reimah). We used the 2019 update WHO-HEAT software for data analysis and performed in steps. First, we disaggregated the IMR disparities by the above-mentioned five equity stratifiers (economic status, education status, infant sex, place of residence and subnational region). Then, we calculated summary measures; Difference (D), Ratio (R), Population Attributable Fraction (PAF) and Absolute Concentration Index (ACI) for each equity stratifiers. However, subnational region inequality in IMR was not calculated for all four summary measures either due to missing or unavailable data in two regions (Mareb and Al-Mhrah). Since ACI is applied only for ordered equity stratifiers such as economic and education status, it was not calculated for place of residence and sex. Details about calculation and interpretations of summary measures are described in the WHO health inequality handbook [18] and technical notes of HEAT software [23]. For some clarity on ACI, it is a complex, weighted measure of inequality that shows the health gradient such as IMR across multiple subgroups with a natural ordering, on an absolute scale and indicates the extent to which IMR was concentrated among the disadvantaged or the advantaged group. If there is no inequality in IMR, ACI takes the value zero. Positive values indicated a concentration of IMR among the advantaged subpopulation (richest and secondary school and above for economic and education status inequality dimensions, respectively), while negative values indicate a concentration of IMR among the disadvantaged subpopulation (poorest and no formal education for economic and education status inequality dimensions, respectively). ACI is usually negative for unfavorable health indicators such as IMR. The larger the absolute value of ACI, the higher the level of inequality of IMR. Difference (D) is a simple, un-weighted measure of inequality that shows the absolute difference between two subgroups. If there was no inequality in IMR, D takes the value of zero. Greater (in absolute values) indicates higher levels of inequality in IMR. A positive value indicated a higher concentration of IMR among infants in the disadvantaged subpopulations (poorest, no formal education, rural residents and male for economic status, education status, place of residence and sex (infant sex) inequality dimensions respectively) and a negative value indicated that higher concentration of IMR among infants in the advantaged subpopulations (richest, secondary school and above, urban residents and female for economic status, education status, place of residence and sex (infant sex) inequality dimensions, respectively). PAF is a complex, weighted measure of inequality that shows the potential for improvement in the national level of IMR, in relative terms, that could be reduced if all subgroups had the same level of IMR as a reference subgroup. PAF was calculated by dividing the population attributable risk (PAR) by the national average µ of IMR and multiplying the fraction by 100. PAF takes negative values for adverse health outcome indicators like IMR. The larger the absolute value of PAF, the larger the degree of inequality. PAF is zero if no further reduction in IMR could be achieved, i.e. if all subgroups have reached the same level of IMR as the reference subgroup. The reference groups are subpopulation with lowest estimate of IMR; richest, secondary school and above, urban residents and female for economic status, education status, place of residence and sex (infant sex) inequality dimensions, respectively. Ratio (R) is a simple, un-weighted measure of inequality that shows the relative inequality between two subgroups. If there is no inequality, R takes the value one. It takes only positive values and the further the value of R from 1, the higher the level of inequality. The demographic health surveys are available publicly and ethics approvals were completed by institutions that commissioned, funded, and managed the surveys. DHS surveys are approved by Inner City Fund (ICF) International and an in-country Institutional Review Board (IRB) to ensure protocols are in compliance with the U.S. Department of Health and Human Services regulations for the protection of human subjects. It is however worth noting that, in this study, we used publicly available DHS data which did not require further ethical clearance.

Materials

N/A

Innovations

The study titled “Socioeconomic, urban‐rural and sex‐based inequality in infant mortality rate: evidence from 2013 Yemen demographic and health survey” provides several recommendations to improve access to maternal health in Yemen. These recommendations can be developed into innovations as follows:

1. Targeted interventions: Develop innovative programs that specifically target the poorest/poor households, rural residents, mothers with no formal education, and male infants. This can include mobile clinics or outreach programs that bring maternal health services directly to these populations, ensuring they have access to prenatal care, skilled birth attendance, and postnatal care.

2. Mobile health (mHealth) solutions: Create innovative mobile applications or text messaging services that provide maternal health information, reminders, and support to women in disadvantaged subpopulations. These solutions can be tailored to the specific needs and challenges faced by these populations, providing guidance on pregnancy care, nutrition, and immunizations.

3. Community-based healthcare services: Innovate community-based healthcare services in rural areas by establishing mobile clinics or community health centers. These facilities can be equipped with trained healthcare providers who can deliver basic prenatal and postnatal care, as well as referrals for more complex cases. Additionally, telemedicine solutions can be implemented to connect healthcare providers in remote areas with specialists in urban areas.

4. Financial support: Develop innovative financial support programs that address the economic barriers faced by disadvantaged households. This can include mobile banking solutions that provide subsidies or vouchers for maternal health services, transportation assistance, and cash transfers to cover the costs associated with accessing healthcare.

5. Education and awareness campaigns: Innovate education and awareness campaigns that utilize various media channels to reach communities with low levels of education. These campaigns can use creative and culturally appropriate methods to raise awareness about the importance of prenatal care, skilled birth attendance, and postnatal care, as well as address any misconceptions or cultural barriers that may exist.

6. Strengthen healthcare infrastructure: Innovate strategies to improve healthcare infrastructure, particularly in rural areas. This can involve the use of telemedicine and telehealth technologies to bridge the gap between rural communities and healthcare facilities, as well as innovative approaches to building and upgrading healthcare facilities in remote areas.

By implementing these innovations, it is possible to improve access to maternal health services and reduce inequalities in infant mortality rate in Yemen.

AI Innovations Description

Based on the study titled “Socioeconomic, urban‐rural and sex‐based inequality in infant mortality rate: evidence from 2013 Yemen demographic and health survey,” the following recommendation can be developed into an innovation to improve access to maternal health:

1. Targeted interventions: To address the observed inequalities in infant mortality rate (IMR), it is important to implement targeted interventions that focus on the poorest/poor households, rural residents, mothers with no formal education, and male infants. These interventions should aim to improve access to maternal health services, including prenatal care, skilled birth attendance, and postnatal care.

2. Mobile health (mHealth) solutions: Utilize mobile technology to provide maternal health information, reminders, and support to women in disadvantaged subpopulations. This can include text messages or mobile applications that provide guidance on pregnancy care, nutrition, and immunizations, as well as reminders for prenatal and postnatal appointments.

3. Community-based healthcare services: Establish community-based healthcare services in rural areas to ensure that women have access to essential maternal health services. This can involve training and deploying community health workers who can provide basic prenatal and postnatal care, as well as referrals for more complex cases.

4. Financial support: Implement financial support programs to alleviate the economic barriers faced by disadvantaged households. This can include providing subsidies or vouchers for maternal health services, transportation assistance, and cash transfers to cover the costs associated with accessing healthcare.

5. Education and awareness campaigns: Conduct education and awareness campaigns to promote the importance of maternal health and encourage women to seek timely and appropriate care. These campaigns should target communities with low levels of education and raise awareness about the benefits of prenatal care, skilled birth attendance, and postnatal care.

6. Strengthen healthcare infrastructure: Invest in improving healthcare infrastructure, particularly in rural areas, by building and upgrading healthcare facilities, ensuring the availability of essential medical equipment and supplies, and training healthcare providers to deliver quality maternal health services.

By implementing these recommendations, it is possible to address the inequalities in access to maternal health and improve maternal and infant health outcomes in Yemen.

AI Innovations Methodology

To simulate the impact of the main recommendations on improving access to maternal health, the following methodology can be employed:

1. Data collection: Collect data on the current status of maternal health access in Yemen, including indicators such as prenatal care coverage, skilled birth attendance, and postnatal care utilization. This data can be obtained from national surveys, health facility records, and other relevant sources.

2. Baseline assessment: Analyze the collected data to determine the existing levels of inequality in maternal health access across different dimensions, such as wealth, education, residence, and infant sex. Calculate summary measures of inequality, such as absolute concentration index (ACI), difference (D), ratio (R), and population attributable fraction (PAF), to quantify the extent of inequality.

3. Intervention implementation: Simulate the implementation of the recommended interventions by applying them to the collected data. For example, target the poorest/poor households, rural residents, mothers with no formal education, and male infants with specific interventions such as targeted interventions, mobile health solutions, community-based healthcare services, financial support programs, and education and awareness campaigns.

4. Impact assessment: Measure the impact of the interventions on improving access to maternal health by comparing the indicators of maternal health access before and after the implementation of the interventions. Calculate the changes in indicators such as prenatal care coverage, skilled birth attendance, and postnatal care utilization for the targeted subpopulations.

5. Inequality analysis: Analyze the impact of the interventions on reducing inequality in maternal health access. Calculate the changes in summary measures of inequality, such as ACI, D, R, and PAF, after the implementation of the interventions. Compare these measures with the baseline assessment to assess the effectiveness of the interventions in reducing inequality.

6. Sensitivity analysis: Conduct sensitivity analysis to assess the robustness of the results. This can involve varying the parameters of the interventions, such as the coverage and effectiveness of targeted interventions or the reach and impact of mobile health solutions, to understand the potential range of outcomes.

7. Policy recommendations: Based on the simulation results, develop policy recommendations for improving access to maternal health in Yemen. These recommendations should be informed by the evidence generated from the simulation and should consider the feasibility, scalability, and sustainability of the interventions.

By following this methodology, policymakers and stakeholders can gain insights into the potential impact of the recommended interventions on improving access to maternal health and reducing inequality in Yemen. This information can guide the development and implementation of effective strategies to address the existing disparities in maternal health access.

Author

Dima Health

Identifiers:

DOI: 10.1186/s13690-021-00589-1

ISSN: 07787367

Research Areas:

Community Interventions, Disparities, Environmental, Food Security, Health System and Policy, Infectious Diseases, Maternal Access, Maternal and Child Health, Quality of Care, Sexual and Reproductive Health, Social Determinants

Study Countries:

Multi-Countries

Study Design:

Cross Sectional Study

Study Approach:

Quantitative

Participants Gender:

Male & Female