Determinants of full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in Ethiopia: A multilevel analysis using 2016 demographic and health survey

Frontiers in Public Health, Volume 11, Year 2023

Interpretation

Background: Despite remarkable improvements in child health services utilization, childhood immunization has been poorly implemented in Ethiopia. However, evidence on the coverage of immunization among children from mothers/caregivers with no education (non-educated mothers were the most identified risk for underutilization of services) are scarce. Therefore, this study aimed to assess the determinants of full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in Ethiopia. Methods: We analyzed data from the 2016 Ethiopia Demographic and Health Survey (EDHS) on a sample of 1,170 children 12–23 months of age identified from deviant mothers/caregivers (mothers/caregivers with no education) through a two-stage stratified sampling. A multilevel mixed-effect binary logistic regression analysis was used to identify the individual and community level determinants of full immunization coverage among children 12–23 months of age with their deviant mothers/caregivers. In the final model, a p-value of < 0.05 and adjusted odds ratio (AOR) with 95% confidence interval (CI) were used to select statistically significant determinants of full immunization coverage. Results: The overall full immunization coverage among children 12–23 months of age identified from deviant mothers/caregivers was 27.4% (95%CI: 25.0, 31.0) in Ethiopia. Deviant mothers/caregivers who are employed (AOR = 1.69, 95%CI: 1.68, 2.45), being in the rich household wealth status (AOR = 2.54, 95%CI: 1.53, 4.22), residing in city (AOR = 5.69, 95%CI: 2.39, 13.61), having one to three (AOR: 3.28, 95% CI: 2.12–5.07) and four and more ANC follow-up during the recent pregnancy (AOR: 3.91, 95% CI: 2.45, 6.24) were the determinants that increased full immunization coverage among children 12–23 months of age. Conclusions: Full immunization coverage among children 12–23 months of age from non-educated mothers/caregivers was low and far behind the national target of coverage. Therefore, a system-wide intervention should be used to enhance employability, wealth status, and key maternal health services like ANC follow-up among non-educated mothers/caregivers to increase their children's full immunization coverage. This analysis was conducted using a cross-sectional data from the EDHS 2016. The EDHS is a nationally representative household survey implemented by the Central Statistical Agency (CSA) of Ethiopia every 5 years (12). Ethiopia was home to more than 120 million people in 2022, of which 16% were children under 5 years (32). Administratively, the country is divided into nine regions [Tigray, Afar, Amhara, Oromia, Benishangul, Gambela, South Nation Nationalities and Peoples' Region (SNNPR), Harari, and Somali] and two City Administrations (Addis Ababa and Dire-Dawa). These nine regions can be divided into developed regions (Tigray, Amhara, Oromia, SNNPR, and Harari) and emerging regions (Afar, Somalia, Benishangul, and Gambela). A developed region and city administrations have a relatively dense population, better infrastructure, education services and better accessibility to health, including immunization services (33). In contrast, in emerging regions, where scattered pastoralists are the majority. It is common for emerging regions to suffer from inadequate infrastructure, inaccessible health services, droughts, poverty, and a lack of clear and detailed regulations (34). The Ethiopian CSA performed a population and housing census in 2007, which was utilized as a sample frame for the 2016 EDHS and provided a complete list of 84,915 enumeration areas (27). To select study participants, the EDHS used a two-stage stratified sampling approach. Each stratum had a sample of EAs, which were chosen at random. Accordingly, all children aged 12–23 months who are regular members of the selected households were eligible for the survey. Finally, a total of 1,170 children 12–23 months of age from deviant mothers/caregivers were identified (Figure 1). Study sample of children 12–23 months of age with their deviant mothers/caregivers in the 2016 Ethiopia DHS (n = 1,170). To identify the positive deviance of mothers/caregivers for full immunization coverage among children aged 12–23 months and determinants of being positive deviant, Anderson's behavioral model of health service (35) and other related studies were used (1, 17, 20, 23, 36). Accordingly, education is the primary determinant of health services utilization. We selected mothers/caregivers with no formal education as a sub-group with a very low likelihood of fully immunizing their children, as mother/caregiver education was the strongest predictor of full immunization coverage after adjusting for the other risk factors associated with full immunization coverage among children in this population. Positive deviant mothers/caregivers were those who reported no formal education but their children fully immunized. Finally, in the analysis, we compared the characteristics of the PD mothers/caregivers to those of their counterparts. Due to significant variations by clusters in the overall full immunization coverage among children aged 12–23 months of age from deviant mothers/caregivers, analysis was stratified by individual and community level. The outcome variable for this study was full vaccination coverage among children 12–23 months of age from deviant mothers/caregiver's which is defined as a child who has had one dose of BCG, three doses of pentavalent, three doses of polio, two doses of Rota, three doses of PCV, and one dose of measles (30). If the child had obtained all of the recommended doses of all vaccines, the immunization status was recoded as “1” and classified as “fully immunized”, or if the child had missed one or more doses, the immunization status was recoded as “0” and was classified as “not fully immunized” (14). Our study assessed independent variables by considering the individual and community-level variables (1, 6, 14, 18, 20, 23). Individual-level variables include, the age of deviant mother/caregivers recoded in completed years (15–24, 25–34, 35+), employment status (employed, non-employed), religion (muslim, orthodox, and other), marital status (married, not married), household wealth status (poor, middle, and rich), head of household (male, female), sex of child (male, female), health insurance coverage (yes, no), i.e., in Ethiopia, community-based health insurance the only health insurance that has been implemented in all regions at household level (37), educational status of husband (no education, primary, secondary, and above), number of ANC visit (no visit, 1–3 visits, 4+ visits), place of delivery (home, health facility), parity (1, 2–5, 6+), childbirth order (1, 2–5, 6+) and uptake of postnatal care (PNC) (yes, no). The uptake of PNC services was assessed whether women received PNC services within 2 months after delivery, regardless of their place of birth. PNC services were assessed based on the mothers/caregiver's verbal responses during the survey. Therefore, it was categorized as “yes” if a woman had at least one PNC visit; otherwise “no.” The wealth index is a composite measure of a household's cumulative living standard. It is calculated using readily available data on a household's ownership of certain assets, such as televisions and bicycles, materials used for housing construction, and types of water access and sanitation facilities. The household wealth index was originally classified into five categories (poorest, poorer, middle, richer, and richest) by the DHS, which was done with principal component analysis (12). However, for analysis in this study, we divided wealth status into three categories: poor, average, and rich. On the other hand, the community-level variables include, place of residence (rural, urban), region (emerging region, developed region, and city administration), the difficulty of getting health services (big problem or not big problem) and media exposure. Deviant mothers/caregiver's media exposure was assessed from the three variables: watching television, listening radio, and reading a newspaper, and labeled as “yes” if a woman has exposure to either of the three media sources at least once a week or “no” if a woman has exposure to none of them. The STATA software version 16 was used to extract, clean, recode, and analyze the data. The descriptive statistics were presented via tables, figures, and narrations. The EDHS data were collected using multistage stratified cluster sampling techniques; as a result, the data had a hierarchical (individuals were nested within communities) nature. Besides, selected and interviewed deviant mothers/caregivers in the same cluster are more likely to be similar to each other than deviant mothers/caregivers from another cluster. This implies that there is a need to consider the between cluster variability by using advanced models. Therefore, to identify determinants, and to estimate the effect of independent variables on full immunization coverage among children 12–23 months of age with their deviant mothers/caregivers, we used the multilevel binary logistic regression analysis method. The Interclass Correlation Coefficient (ICC) and Median Odds Ratio (MOR) were checked to assess whether there was significant clustering or not (38). Accordingly, we found 48% of ICC in our study which showed that 48% of the variation in full coverage among children 12–23 months of age from deviant mothers/caregivers can be explained by clustering. Four models were fitted in this study—null model (no explanatory variables), model I (individual-level factors), model II (community-level factors), model III (both individual and community-level factors). The ICC and deviance (-2* log-likelihood ratio) were used to evaluate model comparison and fitness. Model III was selected as the best-fitted model since it had the lowest deviance. The proportion of variance (PCV) explained by the grouping structure in the population was calculated to analyze the variation between clusters (39). In the bivariable analysis, variables with a p-value < 0.2 were considered for multivariable analysis in each three models. Finally, adjusted odds ratios (AOR) with 95% CI and p-value of ≤ 0.05 in the multivariable analysis were used to declare statistically significant determinants of full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in the final model. Multicollinearity was tested using the variance inflation factor (VIF). There was a VIF of <5 for each independent variable with a mean VIF of 1.85, indicating no significant multicollinearity between independent variables.

AI Digest

Study Justification:

– Despite improvements in child health services utilization, childhood immunization has been poorly implemented in Ethiopia.
– There is a lack of evidence on the coverage of immunization among children from mothers/caregivers with no education, who are at high risk for underutilization of services.
– This study aimed to assess the determinants of full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in Ethiopia.

Highlights:

– The overall full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in Ethiopia was 27.4%.
– Deviant mothers/caregivers who were employed, in the rich household wealth status, residing in the city, and had one to three or four and more ANC follow-up during the recent pregnancy had increased full immunization coverage among their children.
– Full immunization coverage among children from non-educated mothers/caregivers was low and far behind the national target.

Recommendations:

– Implement a system-wide intervention to enhance employability, wealth status, and key maternal health services like ANC follow-up among non-educated mothers/caregivers to increase their children’s full immunization coverage.

Key Role Players:

– Ministry of Health: Responsible for implementing and coordinating immunization programs.
– Health Workers: Provide immunization services and education to mothers/caregivers.
– Community Health Workers: Conduct outreach activities and promote immunization in communities.
– Non-Governmental Organizations: Support immunization programs through advocacy, funding, and implementation.

Cost Items for Planning Recommendations:

– Training and capacity building for health workers and community health workers.
– Outreach activities and community mobilization.
– Development and dissemination of educational materials.
– Monitoring and evaluation of immunization programs.
– Support for ANC services and follow-up.
– Infrastructure and equipment for immunization service delivery.
Please note that the cost items provided are general categories and may vary based on specific program needs and context.

Strength of Evidence

The strength of evidence for this abstract is 7 out of 10.
The evidence in the abstract is rated 7 because it provides detailed information about the study design, data source, sample size, and statistical analysis. However, it does not mention the specific results of the analysis or provide any statistical measures of significance. To improve the evidence, the abstract should include key findings, such as the significant determinants of full immunization coverage, along with their corresponding effect sizes and confidence intervals.

Abstract

This analysis was conducted using a cross-sectional data from the EDHS 2016. The EDHS is a nationally representative household survey implemented by the Central Statistical Agency (CSA) of Ethiopia every 5 years (12). Ethiopia was home to more than 120 million people in 2022, of which 16% were children under 5 years (32). Administratively, the country is divided into nine regions [Tigray, Afar, Amhara, Oromia, Benishangul, Gambela, South Nation Nationalities and Peoples' Region (SNNPR), Harari, and Somali] and two City Administrations (Addis Ababa and Dire-Dawa). These nine regions can be divided into developed regions (Tigray, Amhara, Oromia, SNNPR, and Harari) and emerging regions (Afar, Somalia, Benishangul, and Gambela). A developed region and city administrations have a relatively dense population, better infrastructure, education services and better accessibility to health, including immunization services (33). In contrast, in emerging regions, where scattered pastoralists are the majority. It is common for emerging regions to suffer from inadequate infrastructure, inaccessible health services, droughts, poverty, and a lack of clear and detailed regulations (34). The Ethiopian CSA performed a population and housing census in 2007, which was utilized as a sample frame for the 2016 EDHS and provided a complete list of 84,915 enumeration areas (27). To select study participants, the EDHS used a two-stage stratified sampling approach. Each stratum had a sample of EAs, which were chosen at random. Accordingly, all children aged 12–23 months who are regular members of the selected households were eligible for the survey. Finally, a total of 1,170 children 12–23 months of age from deviant mothers/caregivers were identified (Figure 1). Study sample of children 12–23 months of age with their deviant mothers/caregivers in the 2016 Ethiopia DHS (n = 1,170). To identify the positive deviance of mothers/caregivers for full immunization coverage among children aged 12–23 months and determinants of being positive deviant, Anderson's behavioral model of health service (35) and other related studies were used (1, 17, 20, 23, 36). Accordingly, education is the primary determinant of health services utilization. We selected mothers/caregivers with no formal education as a sub-group with a very low likelihood of fully immunizing their children, as mother/caregiver education was the strongest predictor of full immunization coverage after adjusting for the other risk factors associated with full immunization coverage among children in this population. Positive deviant mothers/caregivers were those who reported no formal education but their children fully immunized. Finally, in the analysis, we compared the characteristics of the PD mothers/caregivers to those of their counterparts. Due to significant variations by clusters in the overall full immunization coverage among children aged 12–23 months of age from deviant mothers/caregivers, analysis was stratified by individual and community level. The outcome variable for this study was full vaccination coverage among children 12–23 months of age from deviant mothers/caregiver's which is defined as a child who has had one dose of BCG, three doses of pentavalent, three doses of polio, two doses of Rota, three doses of PCV, and one dose of measles (30). If the child had obtained all of the recommended doses of all vaccines, the immunization status was recoded as “1” and classified as “fully immunized”, or if the child had missed one or more doses, the immunization status was recoded as “0” and was classified as “not fully immunized” (14). Our study assessed independent variables by considering the individual and community-level variables (1, 6, 14, 18, 20, 23). Individual-level variables include, the age of deviant mother/caregivers recoded in completed years (15–24, 25–34, 35+), employment status (employed, non-employed), religion (muslim, orthodox, and other), marital status (married, not married), household wealth status (poor, middle, and rich), head of household (male, female), sex of child (male, female), health insurance coverage (yes, no), i.e., in Ethiopia, community-based health insurance the only health insurance that has been implemented in all regions at household level (37), educational status of husband (no education, primary, secondary, and above), number of ANC visit (no visit, 1–3 visits, 4+ visits), place of delivery (home, health facility), parity (1, 2–5, 6+), childbirth order (1, 2–5, 6+) and uptake of postnatal care (PNC) (yes, no). The uptake of PNC services was assessed whether women received PNC services within 2 months after delivery, regardless of their place of birth. PNC services were assessed based on the mothers/caregiver's verbal responses during the survey. Therefore, it was categorized as “yes” if a woman had at least one PNC visit; otherwise “no.” The wealth index is a composite measure of a household's cumulative living standard. It is calculated using readily available data on a household's ownership of certain assets, such as televisions and bicycles, materials used for housing construction, and types of water access and sanitation facilities. The household wealth index was originally classified into five categories (poorest, poorer, middle, richer, and richest) by the DHS, which was done with principal component analysis (12). However, for analysis in this study, we divided wealth status into three categories: poor, average, and rich. On the other hand, the community-level variables include, place of residence (rural, urban), region (emerging region, developed region, and city administration), the difficulty of getting health services (big problem or not big problem) and media exposure. Deviant mothers/caregiver's media exposure was assessed from the three variables: watching television, listening radio, and reading a newspaper, and labeled as “yes” if a woman has exposure to either of the three media sources at least once a week or “no” if a woman has exposure to none of them. The STATA software version 16 was used to extract, clean, recode, and analyze the data. The descriptive statistics were presented via tables, figures, and narrations. The EDHS data were collected using multistage stratified cluster sampling techniques; as a result, the data had a hierarchical (individuals were nested within communities) nature. Besides, selected and interviewed deviant mothers/caregivers in the same cluster are more likely to be similar to each other than deviant mothers/caregivers from another cluster. This implies that there is a need to consider the between cluster variability by using advanced models. Therefore, to identify determinants, and to estimate the effect of independent variables on full immunization coverage among children 12–23 months of age with their deviant mothers/caregivers, we used the multilevel binary logistic regression analysis method. The Interclass Correlation Coefficient (ICC) and Median Odds Ratio (MOR) were checked to assess whether there was significant clustering or not (38). Accordingly, we found 48% of ICC in our study which showed that 48% of the variation in full coverage among children 12–23 months of age from deviant mothers/caregivers can be explained by clustering. Four models were fitted in this study—null model (no explanatory variables), model I (individual-level factors), model II (community-level factors), model III (both individual and community-level factors). The ICC and deviance (-2* log-likelihood ratio) were used to evaluate model comparison and fitness. Model III was selected as the best-fitted model since it had the lowest deviance. The proportion of variance (PCV) explained by the grouping structure in the population was calculated to analyze the variation between clusters (39). In the bivariable analysis, variables with a p-value < 0.2 were considered for multivariable analysis in each three models. Finally, adjusted odds ratios (AOR) with 95% CI and p-value of ≤ 0.05 in the multivariable analysis were used to declare statistically significant determinants of full immunization coverage among children 12–23 months of age from deviant mothers/caregivers in the final model. Multicollinearity was tested using the variance inflation factor (VIF). There was a VIF of <5 for each independent variable with a mean VIF of 1.85, indicating no significant multicollinearity between independent variables.

Materials

N/A

Innovations

Based on the provided analysis, here are some potential innovations that could improve access to maternal health:

1. Mobile Health (mHealth) Interventions: Develop mobile applications or text messaging services to provide information and reminders about immunization schedules, antenatal care visits, and postnatal care. This can help educate and empower mothers/caregivers, especially those with no formal education, to stay on track with their child’s immunization and healthcare needs.

2. Community Health Workers (CHWs): Train and deploy CHWs in rural and underserved areas to provide education, counseling, and support to mothers/caregivers. CHWs can conduct home visits, organize community immunization campaigns, and address any concerns or barriers to accessing maternal health services.

3. Transportation Support: Establish transportation services or vouchers to help mothers/caregivers overcome geographical barriers and reach healthcare facilities for antenatal care, immunization appointments, and postnatal care. This can ensure that even those in remote areas have access to essential maternal health services.

4. Financial Incentives: Implement financial incentives, such as conditional cash transfers or vouchers, to encourage mothers/caregivers to attend antenatal care visits, complete immunization schedules, and seek postnatal care. This can help alleviate financial barriers and increase utilization of maternal health services.

5. Health Education Programs: Develop targeted health education programs that specifically address the needs and concerns of mothers/caregivers with no formal education. These programs should use simple language, visual aids, and interactive methods to effectively communicate important maternal health information and promote behavior change.

6. Strengthening Health Systems: Invest in improving the overall health system infrastructure, including healthcare facilities, staffing, and supply chains. This will ensure that maternal health services, including immunization, are readily available, accessible, and of high quality for all mothers/caregivers, regardless of their educational background.

It is important to note that these recommendations are based on the specific context and findings of the analysis provided. Further research and contextual adaptation may be necessary to effectively implement these innovations in Ethiopia or other similar settings.

AI Innovations Description

The analysis conducted in this study aimed to identify the determinants of full immunization coverage among children 12-23 months of age from deviant mothers/caregivers in Ethiopia. The study used data from the 2016 Ethiopia Demographic and Health Survey (EDHS) and employed a multilevel mixed-effect binary logistic regression analysis.

The findings of the study revealed that the overall full immunization coverage among children 12-23 months of age from deviant mothers/caregivers in Ethiopia was 27.4%. Several determinants were identified that increased the likelihood of full immunization coverage. These determinants included:

1. Employment status: Deviant mothers/caregivers who were employed had a higher likelihood of their children being fully immunized.

2. Household wealth status: Children from deviant mothers/caregivers in the rich household wealth status had a higher likelihood of full immunization coverage.

3. Place of residence: Children residing in cities had a significantly higher likelihood of full immunization coverage compared to those in rural areas.

4. Antenatal care (ANC) follow-up: Deviant mothers/caregivers who had one to three or four and more ANC follow-up visits during their recent pregnancy had a higher likelihood of their children being fully immunized.

Based on these findings, the study concluded that full immunization coverage among children 12-23 months of age from non-educated mothers/caregivers was low and below the national target. To improve access to maternal health and increase full immunization coverage, the study recommended a system-wide intervention. This intervention should focus on enhancing employability, improving household wealth status, and promoting key maternal health services such as ANC follow-up among non-educated mothers/caregivers.

It is important to note that the study utilized cross-sectional data from the EDHS 2016, which is a nationally representative household survey conducted every 5 years in Ethiopia. The study sample included 1,170 children 12-23 months of age from deviant mothers/caregivers, selected through a two-stage stratified sampling approach. The analysis considered individual-level and community-level variables to identify the determinants of full immunization coverage.

Overall, the study provides valuable insights into the factors influencing full immunization coverage among children from deviant mothers/caregivers in Ethiopia and highlights the need for targeted interventions to improve access to maternal health services and increase immunization coverage.

AI Innovations Methodology

Based on the provided information, here are some potential recommendations to improve access to maternal health:

1. Enhance maternal education: Implement programs that focus on increasing educational opportunities for mothers/caregivers with no formal education. This can be done through adult education programs, vocational training, and awareness campaigns highlighting the importance of education for maternal and child health.

2. Strengthen employment opportunities: Develop initiatives that promote employment opportunities for mothers/caregivers. This can include skill-building programs, job placement services, and support for entrepreneurship. By improving economic stability, mothers/caregivers will have better access to healthcare services, including maternal health.

3. Improve household wealth status: Implement strategies to address poverty and income disparities among households. This can involve providing financial assistance, microfinance programs, and income-generating activities to uplift families out of poverty. Increased household wealth can contribute to better access to healthcare services, including maternal health.

4. Enhance antenatal care (ANC) services: Focus on increasing the number of ANC visits among mothers/caregivers. This can be achieved through community outreach programs, mobile clinics, and awareness campaigns that emphasize the importance of regular ANC visits. By ensuring adequate ANC, mothers/caregivers can receive necessary vaccinations and healthcare services during pregnancy.

5. Strengthen health infrastructure in rural and emerging regions: Allocate resources to improve healthcare infrastructure, particularly in rural and emerging regions. This can involve building and upgrading health facilities, ensuring the availability of trained healthcare professionals, and improving transportation networks to facilitate access to maternal health services.

To simulate the impact of these recommendations on improving access to maternal health, a methodology could be developed as follows:

1. Define indicators: Identify key indicators that measure access to maternal health, such as the percentage of mothers/caregivers receiving ANC, the percentage of mothers/caregivers receiving vaccinations, and the percentage of mothers/caregivers with access to skilled birth attendants.

2. Collect baseline data: Gather data on the current status of these indicators in the target population. This can be done through surveys, interviews, and data from healthcare facilities and government agencies.

3. Develop a simulation model: Create a simulation model that incorporates the identified recommendations and their potential impact on the selected indicators. This model should consider factors such as population size, demographic characteristics, and geographical distribution.

4. Input data and run simulations: Input the baseline data into the simulation model and run multiple simulations to assess the potential impact of the recommendations. Vary the input parameters to account for different scenarios and assumptions.

5. Analyze results: Analyze the simulation results to determine the projected changes in the selected indicators. Assess the effectiveness of each recommendation and identify any potential trade-offs or unintended consequences.

6. Refine and validate the model: Refine the simulation model based on the analysis of results and feedback from stakeholders. Validate the model by comparing the simulated outcomes with real-world data and expert opinions.

7. Communicate findings and make recommendations: Present the simulation findings to relevant stakeholders, policymakers, and healthcare professionals. Use the results to inform decision-making and advocate for the implementation of the recommended interventions.

By following this methodology, policymakers and healthcare providers can gain insights into the potential impact of different interventions on improving access to maternal health. This can guide resource allocation, policy development, and program implementation to effectively address the identified challenges.

Authors & Co-Authors

Statistics:

Authors: 10

Identifiers:

DOI: 10.3389/fpubh.2023.1085279

Research Areas:

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

Study Countries:

Ethiopia, Multi-Countries, Somalia

Study Design:

Case-Control Study, Cohort Study, Cross Sectional Study, randomised-control-trial

Study Approach:

Quantitative

Participants Gender:

Male & Female