A multilevel analysis of factors associated with teenage pregnancy in ethiopia

International Journal of Women's Health, Volume 12, Year 2020

Interprétation

Background: A significant number of girls in Ethiopia begin childbearing at an early age. Teenage pregnancy is the main contributor to maternal and child morbidity and mortality, and the vicious cycle of ill-health and poverty. However limited evidence exists about individual-and community-level factors affecting teenage pregnancy in Ethiopia. Methods: This study used data from the 2016 Ethiopian Demographic and Health Survey (EDHS). A total of 3381 (weighted) teenagers aged 15–19 years were included in the study. A two-stage stratified cluster was used. Data were analyzed using Stata version 14. Multilevel mixed effect logistic regression was used to identify factors affecting teenage pregnancy. Results: Being 17 (AOR=9.26, 95% CI=2.67–32.04), 18 (AOR=9.53, 95% CI=2.97–30.04) and 19 years old (AOR=20.01, 95% CI=5.94–67.39), uneducated (AOR=3.83, 95% CI=1.05–14.00), primary educated (AOR=3.34, 95% CI=1.01–11.08), being married (AOR=70.12, 95% CI=27.55–178.4), and communities with a higher proportion of poor (AOR=3.86, 95% CI=1.80–8.26) were predictors of teenage pregnancy. Conclusion: Age, educational status, and marital status from individual-level factors, and community wealth status from community-level factors were predictors of teenage pregnancy. The government should strive to improve female education, and fight against early marriage and sexual initiation. The study was conducted in Ethiopia, which is one of the Sub-Saharan African countries. It is found in the North-Eastern part of Africa, lies between 3° and 15° North latitude and 33° and 48° East longitude.37 It has a total estimated 114,530,078 population. Females’ age 15–19 years old are estimated to total 6.3 million.38 Ethiopia is one of the poorest counties, with a gross domestic product (GDP) per capita income of US$772. Nearly one-fourth of the populations of Ethiopia are living below the national poverty line.39 Though Ethiopia is making the fastest progress in ensuring access to education in SSA, it still faces challenges; low primary completion rates, a fall in enrollment rates in secondary education (30.7%), and low-quality education at all levels.40 This study was an in-depth secondary analysis of the 2016 EDHS. The EDHS has been conducted every 5 years to provide health and health-related indicators in Ethiopia. The 2016 EDHS is the latest national survey conducted in nine regional states and two administrative cities. The regions include Tigray, Afar, Amhara, Oromia, Somali, Benishangul-Gumuz, Southern Nations, Nationalities and Peoples’ Region (SNNPR), (Gambella, and Harari. Administrative cities include Addis Ababa and Dire Dawa). Administratively, regions in Ethiopia are divided into Zones, and Zones into administrative units called Woredas. Each Woreda is further subdivided into the lowest administrative unit, called Kebeles. During the 2007 census, each Kebele was subdivided into census enumeration areas (EAs), which were convenient for the implementation of the census.7 The 2016 EDHS was cross-sectional by design. The 2016 EDHS sample was stratified and selected in two stages. In the first stage, stratification was conducted by region and then in each region stratified as urban and rural, yielding 21 sampling strata. A total of 645 EAs (202 in urban areas and 443 in rural areas) were selected with probability proportional to EA size in each sampling stratum. In the second stage, a fixed number of 28 households per cluster were with an equal probability systematic selection from the newly created household listing. The outcome variable was dichotomized as teenage pregnancy (yes/no). A woman was considered as experiencing teenage pregnancy if her age was from 15–19 and if she had ever been pregnant before or during the survey. Regarding media exposure, a woman was considered as having media exposure if she listened to both radio and television at least once a week. The variable wealth index was re-categorized as “Poor”, “Middle”, and “Rich” categories by merging poorest with poorer and richest with richer.25 Community-level variables were computed by aggregating the individual women characteristics into clusters. Then the proportion was calculated by dividing subcategories to the total. Distributions of the proportion of aggregate variables were checked using the Shapiro–Wilk normality test and were not normally distributed. Therefore, these aggregate variables were categorized using the median value. A total of eight community variables were generated. Family disruption was created using the proportion of female-headed houses in their cluster.33 Community educational status was computed based on the proportion of below secondary educational status in their cluster. Community wealth status was computed using the proportion of poor wealth index in each cluster. Community-level literacy was calculated based on the proportion illiterate in each cluster; the same is true for other community-level variables.25 Data were cleaned to check its consistency and missing values. Descriptive statistics such as frequencies, median, and percentages were computed. The data were analyzed using Stata version 14.0. Sampling weights were done to compensate for the non-proportional allocation of the sample to strata as well as for non-responses. The EDHS data are hierarchical, ie, individuals were nested within communities, and Intra-class Correlation Coefficient (ICC) was greater than 10% (ICC=34%). Therefore, a two-level mixed-effects logistic regression model was conducted to estimate both independent (fixed) effects of the explanatory variables and community-level random effects on teenage pregnancy. The log of the probability of being pregnant at teenage was modelled using a two-level multilevel model as follows;41 Where, πij is the probability of being pregnant for the ith teenager in the jth community; 1-πij is the probability of being a non-pregnant teenager; i and j are the level 1 (individual) and level 2 (community) units, respectively; X and Z refer to individual- and community-level variables, respectively; the β’s are the fixed coefficients –therefore, for every one-unit increase in X/Z there is a corresponding effect on the probability of being pregnant as a teenager. Whereas, β0 is the intercept – the effect on the probability of being pregnant as a teenager in the absence of influence of predictors; and uj shows the random effect (effect of the community on a teenager to become pregnant) for the jth community and eij showed random errors at the individual levels. By assuming each community had a different intercept (β0) and fixed coefficient (β), the clustered data nature and the within and between community variations were taken into account. In the analysis first, bivariable multilevel logistic regression was computed and variables with a P-value less than 0.3 were included in multivariable multilevel logistic regression. Four models were displayed in this analysis, Model 0 (model containing no factors), Model 1 (containing only individual factors), Model 2 (containing only community factors), and Model 3 (both individual- and community-level factors). Variables with a P-value of less than 0.05 had a statistically significant association with the outcome variable. The result of the fixed effect was presented as Adjusted Odds Ratio (AOR) with their 95% confidence interval (95% CI). The measures of variation (random-effects) were reported using ICC, a proportional change in variance (PCV), and Median Odds Ratio (MOR). The ICC was used to show how much the observation within one cluster resembled each other, and MOR is a measure of unexplained cluster heterogeneity. The ICC was computed using this formula as follows: [], where is the estimated variance of clusters. MOR is the median value of the odds ratio between the area at highest risk and the area at the lowest risk when randomly picking out two areas and calculated using the formula []. The proportional change in variance (PCV) signifies the total variation attributed by individual-level factors and area-level factors in the multilevel model. Standard error at the cut-off point of ±2 was used to check multicollinearity and there was no multicollinearity. The goodness of fit of the model was checked by the log-likelihood test. An authorization letter was also obtained from CSA for downloading the EDHS data set by requesting the website www.measuredhs.com. Ethical clearance for the primary study (EDHS 2016) was obtained from Ethiopia Health and Nutrition Research Institute Review Board, the Ministry of Science and Technology, Institutional Review Board of ICF International, and the Centers of Disease Control and Prevention (CDC). The accessed data were used for the registered research only. All data were treated as confidential and no effort was made to identify any household or individual respondent interviewed in the survey. The detailed information on methodology and the ethical issue was published in the EDHS report.

Résumé de l’IA

Study Justification:

– Teenage pregnancy is a significant issue in Ethiopia, contributing to maternal and child morbidity and mortality, as well as perpetuating ill-health and poverty.
– Limited evidence exists on the individual and community factors that affect teenage pregnancy in Ethiopia.
– Understanding these factors is crucial for developing effective interventions and policies to address teenage pregnancy.

Study Highlights:

– The study used data from the 2016 Ethiopian Demographic and Health Survey (EDHS) and included 3381 teenagers aged 15-19 years.
– Factors associated with teenage pregnancy were identified using multilevel mixed effect logistic regression analysis.
– The study found that age, educational status, marital status, and community wealth status were predictors of teenage pregnancy.
– Recommendations were made to improve female education, fight against early marriage, and address sexual initiation.

Recommendations for Lay Reader and Policy Maker:

– Improve female education: Efforts should be made to increase access to and quality of education for girls, with a focus on reducing dropout rates and ensuring completion of primary and secondary education.
– Fight against early marriage: Policies and interventions should be implemented to discourage early marriage and promote age-appropriate relationships.
– Address sexual initiation: Comprehensive sexual and reproductive health education programs should be developed and implemented to provide accurate information and promote responsible sexual behavior among teenagers.

Key Role Players:

– Ministry of Education: Responsible for implementing policies and programs to improve female education.
– Ministry of Women, Children, and Youth Affairs: Involved in efforts to combat early marriage and promote gender equality.
– Ministry of Health: Responsible for implementing sexual and reproductive health programs and providing access to contraceptives and family planning services.
– Non-governmental organizations (NGOs): Play a crucial role in implementing interventions and providing support services to address teenage pregnancy.

Cost Items for Planning Recommendations:

– Education infrastructure: Investment in schools, classrooms, and educational materials to improve access and quality of education.
– Teacher training: Funding for training programs to enhance the skills and knowledge of teachers in delivering comprehensive sexual and reproductive health education.
– Awareness campaigns: Budget for public awareness campaigns to promote the importance of education, discourage early marriage, and raise awareness about responsible sexual behavior.
– Health services: Allocation of resources to ensure availability and accessibility of sexual and reproductive health services, including contraceptives and family planning.
– Monitoring and evaluation: Funding for monitoring and evaluation activities to assess the effectiveness of interventions and track progress in reducing teenage pregnancy rates.

Force de la preuve

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 sample size (3381 teenagers) and uses data from the 2016 Ethiopian Demographic and Health Survey. The study also employs a multilevel mixed effect logistic regression analysis to identify factors affecting teenage pregnancy. However, to improve the evidence, the abstract could provide more information on the representativeness of the sample and the response rate of the survey. Additionally, it would be helpful to include the statistical significance of the associations found in the study.

Abstrait

The study was conducted in Ethiopia, which is one of the Sub-Saharan African countries. It is found in the North-Eastern part of Africa, lies between 3° and 15° North latitude and 33° and 48° East longitude.37 It has a total estimated 114,530,078 population. Females’ age 15–19 years old are estimated to total 6.3 million.38 Ethiopia is one of the poorest counties, with a gross domestic product (GDP) per capita income of US$772. Nearly one-fourth of the populations of Ethiopia are living below the national poverty line.39 Though Ethiopia is making the fastest progress in ensuring access to education in SSA, it still faces challenges; low primary completion rates, a fall in enrollment rates in secondary education (30.7%), and low-quality education at all levels.40 This study was an in-depth secondary analysis of the 2016 EDHS. The EDHS has been conducted every 5 years to provide health and health-related indicators in Ethiopia. The 2016 EDHS is the latest national survey conducted in nine regional states and two administrative cities. The regions include Tigray, Afar, Amhara, Oromia, Somali, Benishangul-Gumuz, Southern Nations, Nationalities and Peoples’ Region (SNNPR), (Gambella, and Harari. Administrative cities include Addis Ababa and Dire Dawa). Administratively, regions in Ethiopia are divided into Zones, and Zones into administrative units called Woredas. Each Woreda is further subdivided into the lowest administrative unit, called Kebeles. During the 2007 census, each Kebele was subdivided into census enumeration areas (EAs), which were convenient for the implementation of the census.7 The 2016 EDHS was cross-sectional by design. The 2016 EDHS sample was stratified and selected in two stages. In the first stage, stratification was conducted by region and then in each region stratified as urban and rural, yielding 21 sampling strata. A total of 645 EAs (202 in urban areas and 443 in rural areas) were selected with probability proportional to EA size in each sampling stratum. In the second stage, a fixed number of 28 households per cluster were with an equal probability systematic selection from the newly created household listing. The outcome variable was dichotomized as teenage pregnancy (yes/no). A woman was considered as experiencing teenage pregnancy if her age was from 15–19 and if she had ever been pregnant before or during the survey. Regarding media exposure, a woman was considered as having media exposure if she listened to both radio and television at least once a week. The variable wealth index was re-categorized as “Poor”, “Middle”, and “Rich” categories by merging poorest with poorer and richest with richer.25 Community-level variables were computed by aggregating the individual women characteristics into clusters. Then the proportion was calculated by dividing subcategories to the total. Distributions of the proportion of aggregate variables were checked using the Shapiro–Wilk normality test and were not normally distributed. Therefore, these aggregate variables were categorized using the median value. A total of eight community variables were generated. Family disruption was created using the proportion of female-headed houses in their cluster.33 Community educational status was computed based on the proportion of below secondary educational status in their cluster. Community wealth status was computed using the proportion of poor wealth index in each cluster. Community-level literacy was calculated based on the proportion illiterate in each cluster; the same is true for other community-level variables.25 Data were cleaned to check its consistency and missing values. Descriptive statistics such as frequencies, median, and percentages were computed. The data were analyzed using Stata version 14.0. Sampling weights were done to compensate for the non-proportional allocation of the sample to strata as well as for non-responses. The EDHS data are hierarchical, ie, individuals were nested within communities, and Intra-class Correlation Coefficient (ICC) was greater than 10% (ICC=34%). Therefore, a two-level mixed-effects logistic regression model was conducted to estimate both independent (fixed) effects of the explanatory variables and community-level random effects on teenage pregnancy. The log of the probability of being pregnant at teenage was modelled using a two-level multilevel model as follows;41 Where, πij is the probability of being pregnant for the ith teenager in the jth community; 1-πij is the probability of being a non-pregnant teenager; i and j are the level 1 (individual) and level 2 (community) units, respectively; X and Z refer to individual- and community-level variables, respectively; the β’s are the fixed coefficients –therefore, for every one-unit increase in X/Z there is a corresponding effect on the probability of being pregnant as a teenager. Whereas, β0 is the intercept – the effect on the probability of being pregnant as a teenager in the absence of influence of predictors; and uj shows the random effect (effect of the community on a teenager to become pregnant) for the jth community and eij showed random errors at the individual levels. By assuming each community had a different intercept (β0) and fixed coefficient (β), the clustered data nature and the within and between community variations were taken into account. In the analysis first, bivariable multilevel logistic regression was computed and variables with a P-value less than 0.3 were included in multivariable multilevel logistic regression. Four models were displayed in this analysis, Model 0 (model containing no factors), Model 1 (containing only individual factors), Model 2 (containing only community factors), and Model 3 (both individual- and community-level factors). Variables with a P-value of less than 0.05 had a statistically significant association with the outcome variable. The result of the fixed effect was presented as Adjusted Odds Ratio (AOR) with their 95% confidence interval (95% CI). The measures of variation (random-effects) were reported using ICC, a proportional change in variance (PCV), and Median Odds Ratio (MOR). The ICC was used to show how much the observation within one cluster resembled each other, and MOR is a measure of unexplained cluster heterogeneity. The ICC was computed using this formula as follows: [], where is the estimated variance of clusters. MOR is the median value of the odds ratio between the area at highest risk and the area at the lowest risk when randomly picking out two areas and calculated using the formula []. The proportional change in variance (PCV) signifies the total variation attributed by individual-level factors and area-level factors in the multilevel model. Standard error at the cut-off point of ±2 was used to check multicollinearity and there was no multicollinearity. The goodness of fit of the model was checked by the log-likelihood test. An authorization letter was also obtained from CSA for downloading the EDHS data set by requesting the website www.measuredhs.com. Ethical clearance for the primary study (EDHS 2016) was obtained from Ethiopia Health and Nutrition Research Institute Review Board, the Ministry of Science and Technology, Institutional Review Board of ICF International, and the Centers of Disease Control and Prevention (CDC). The accessed data were used for the registered research only. All data were treated as confidential and no effort was made to identify any household or individual respondent interviewed in the survey. The detailed information on methodology and the ethical issue was published in the EDHS report.

Matériaux

N/A

Innovations

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

1. Mobile Health (mHealth) Solutions: Develop mobile applications or text messaging services that provide pregnant teenagers with information on prenatal care, nutrition, and safe delivery practices. These tools can also send reminders for antenatal visits and provide access to emergency helplines.

2. Community Health Workers: Train and deploy community health workers to provide education and support to pregnant teenagers in their communities. These workers can conduct home visits, provide counseling, and connect teenagers with appropriate healthcare services.

3. School-Based Health Programs: Implement comprehensive sexual and reproductive health education programs in schools to empower teenagers with knowledge about contraception, family planning, and the importance of delaying pregnancy. These programs can also provide access to reproductive health services within school premises.

4. Telemedicine: Establish telemedicine services that allow pregnant teenagers in remote areas to consult with healthcare professionals via video calls or phone calls. This can help overcome geographical barriers and ensure timely access to medical advice and support.

5. Financial Incentives: Introduce financial incentives, such as conditional cash transfers or scholarships, to encourage pregnant teenagers to stay in school and complete their education. This can help reduce the prevalence of early marriages and improve future opportunities for these young mothers.

6. Strengthening Health Infrastructure: Invest in improving healthcare infrastructure, particularly in rural areas, by building and equipping maternal health clinics and hospitals. This can ensure that pregnant teenagers have access to quality prenatal care, safe delivery services, and emergency obstetric care when needed.

7. Public Awareness Campaigns: Launch public awareness campaigns to challenge societal norms and attitudes that contribute to teenage pregnancy. These campaigns can promote gender equality, empower girls, and educate communities about the negative consequences of early pregnancies.

It is important to note that the implementation of these innovations should be context-specific and involve collaboration between government agencies, healthcare providers, NGOs, and local communities.

AI Innovations Description

Based on the study titled “A multilevel analysis of factors associated with teenage pregnancy in Ethiopia,” the following recommendation can be developed into an innovation to improve access to maternal health:

1. Improve female education: The study found that uneducated and primary educated teenagers were more likely to experience teenage pregnancy. Therefore, implementing innovative educational programs targeted at girls, especially in rural areas, can help delay childbearing and empower them to make informed decisions about their reproductive health.

2. Fight against early marriage: The study identified being married as a significant predictor of teenage pregnancy. To address this issue, innovative interventions should focus on raising awareness about the negative consequences of early marriage and promoting alternative pathways for girls, such as vocational training and higher education.

3. Enhance access to sexual and reproductive health services: Innovative approaches should be developed to increase access to comprehensive sexual and reproductive health services for teenagers. This can include mobile health clinics, telemedicine, and community-based outreach programs that provide information, counseling, contraceptives, and antenatal care services.

4. Address community-level factors: The study found that communities with a higher proportion of poor individuals were more likely to have teenage pregnancies. Innovative solutions should aim to address poverty at the community level through economic empowerment programs, job creation initiatives, and social safety nets.

5. Strengthen data collection and analysis: Innovations in data collection and analysis can help identify trends, monitor progress, and inform evidence-based decision-making. This can include the use of digital health technologies, data visualization tools, and predictive analytics to identify high-risk areas and target interventions effectively.

By implementing these recommendations as innovative solutions, access to maternal health can be improved, leading to a reduction in teenage pregnancies and better health outcomes for both mothers and children in Ethiopia.

AI Innovations Methodology

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

1. Increase access to education: Promote and invest in initiatives that improve female education, particularly in rural areas where access to education is limited. This can help delay the age at which girls start childbearing and reduce the risk of teenage pregnancy.

2. Address early marriage: Implement policies and programs that discourage early marriage and promote the legal age of marriage. This can help reduce the incidence of teenage pregnancy and improve maternal health outcomes.

3. Enhance sexual and reproductive health education: Develop comprehensive sexual and reproductive health education programs that provide accurate information on contraception, family planning, and safe sex practices. This can empower young girls to make informed decisions about their reproductive health and reduce the risk of unintended pregnancies.

4. Improve access to contraceptives: Strengthen the availability and accessibility of contraceptives, including long-acting reversible contraceptives (LARCs), in both urban and rural areas. This can help prevent unintended pregnancies and reduce the demand for unsafe abortions.

To simulate the impact of these recommendations on improving access to maternal health, a methodology could include the following steps:

1. Data collection: Gather relevant data on teenage pregnancy rates, maternal health indicators, education levels, early marriage rates, and access to contraceptives in the target population.

2. Baseline assessment: Analyze the current situation and identify the key factors influencing access to maternal health, including teenage pregnancy rates and barriers to education and contraception.

3. Scenario development: Develop different scenarios based on the potential recommendations mentioned above. For each scenario, estimate the expected changes in teenage pregnancy rates, education levels, early marriage rates, and contraceptive use.

4. Modeling and simulation: Use statistical modeling techniques, such as multilevel mixed effect logistic regression, to simulate the impact of the different scenarios on improving access to maternal health. This can help estimate the potential reduction in teenage pregnancy rates and improvements in maternal health outcomes.

5. Sensitivity analysis: Conduct sensitivity analysis to assess the robustness of the results and identify the key drivers of change. This can help policymakers understand the potential uncertainties and limitations of the simulation results.

6. Policy recommendations: Based on the simulation results, provide evidence-based policy recommendations to stakeholders, including government agencies, NGOs, and healthcare providers, to guide interventions and investments aimed at improving access to maternal health.

It is important to note that the methodology described above is a general framework and may need to be adapted based on the specific context and available data in Ethiopia.

Auteur

Dima Health

Statistics:

Citations: 5

Identifiers:

DOI: 10.2147/IJWH.S265201

Research Areas:

Community Interventions, Food Security, Genetics and Genomics, Health System and Policy, Maternal Access, Maternal and Child Health, Quality of Care, Sexual and Reproductive Health, Social Determinants, Technology and Innovations

Study Countries:

Ethiopia

Study Design:

Case-Control Study, Cross Sectional Study

Study Approach:

Mixed-methods, Qualitative, Quantitative

Participants Gender:

Female

Partagez ceci :