Prevalence and determinants of unintended pregnancy in Ethiopia: narrative synthesis and meta-analysis

Determinants of unintended pregnancy and induced abortion among adolescent women in Ethiopia: Evidence from multilevel mixed-effects decomposition analysis of 2000-2016 Ethiopian demographic and health survey data

BACKGROUND

Adolescents are highly at risk of unintended pregnancy due to physiological, sexual, social and psychological growth. The pregnancy may end with early childbirth, induced abortion and its complications. Although, the trends of unintended pregnancy and induced abortion have declined over time in Ethiopia, evidence is limited on key determinants for decline in order to propose vital areas of interventions. The current study aimed to identify the determinants of unintended pregnancy and induced abortion among adolescents over the decades.

METHODS

Trends in the prevalence of unintended pregnancy and induced abortion among adolescent women aged 15-19 years were investigated based using a series of the Ethiopia Demographic and Health Surveys (EDHS) data for the years 2000, 2005, 2011, and 2016. Sub-sample of adolescent women data was extracted from each survey. The combined datasets for unintended pregnancy and induced abortion over the study period (2000-2016) was analyzed. The percentage changes of trends of unintended pregnancy and induced abortion with its corresponding 95% CI for each variable were calculated. Multilevel mixed-effects decomposition analysis was applied to identify factors significantly associated with trends of unintended pregnancy and induced abortion among adolescents.

RESULTS

The trends of unintended pregnancy and induced abortion significantly declined during the study period. Unintended pregnancy among Ethiopian adolescents aged 15-19 years significantly decreased from 307 (41.4%) (95% CI: 35.7, 47.2%, p<0.001) in 2000 to 120 (25.1%) (95% CI: 18.9, 31.4%) in 2016. On the other hand, induced abortion significantly decreased from 62 (8.3%) (95% CI: 5.2, 11.4%) in 2000 to 20 (4.1%) (95% CI: 1.3, 6.9%, p = 0.004) in 2016. Age older than 18 years (Coeff = -0.41, 95%CI, -0.64, -0.18, p<0.001), living in Somali regional state (Coeff = -2.21, 95%CI, -3.27, -1.15, p<0.001) and exposure to media (Coeff = -0.60, 95%CI, -0.87, -0.33, p<0.001) showed a significance association with decline in unintended pregnancy whereas; living in Benshangul-Gumuz regional state (Coeff = -0.17, 95%CI, -0.32, -0.19, p = 0.03) and ANC service utilization history (Coeff = -0.81, 95%CI, -1.45, -0.17, p = 0.01) showed significance association with decline in induced abortion.

CONCLUSION

The trends of unintended pregnancy and induced abortion significantly declined over the past decades in Ethiopia. Adolescent girls aged 17 years and above, exposure to media and living in Somali showed significant association with decline in unintended pregnancy whereas; living in Benshangul-Gumuz and ANC service utilization history showed significant decline with induced abortion. Exposure to media and utilization of Antenatal care (ANC) services may improve adolescent girls' reproductive health uptake.

Machine learning to predict unintended pregnancy among reproductive-age women in Ethiopia: evidence from EDHS 2016

BACKGROUND

An unintended pregnancy is a pregnancy that is either unwanted or mistimed, such as when it occurs earlier than desired. It is one of the most important issues the public health system is currently facing, and it comes at a significant cost to society both economically and socially. The burden of an undesired pregnancy still weighs heavily on Ethiopia. The purpose of this study was to assess the effectiveness of machine learning algorithms in predicting unintended pregnancy in Ethiopia and to identify the key predictors.

METHOD

Machine learning techniques were used in the study to analyze secondary data from the 2016 Ethiopian Demographic and Health Survey. To predict and identify significant determinants of unintended pregnancy using Python software, six machine-learning algorithms were applied to a total sample of 7193 women. The top unplanned pregnancy predictors were chosen using the feature importance technique. The effectiveness of such models was evaluated using sensitivity, specificity, accuracy, and area under the curve.

RESULT

The ExtraTrees classifier was chosen as the top machine learning model after various performance evaluations. The region, the ideal number of children, religion, wealth index, age at first sex, husband education, refusal sex, total births, age at first birth, and mother's educational status are identified as contributing factors in that predict unintended pregnancy.

CONCLUSION

The ExtraTrees machine learning model has a better predictive performance for identifying predictors of unintended pregnancies among the chosen algorithms and could improve with better policy decision-making in this area. Using these important features to help direct appropriate policy can significantly increase the chances of mother survival.

The factors associated with under-five mortality in Ethiopia

Background

Under-five mortality refers to the likelihood of dying between the ages of birth and five. The number of children under the age of five who die each year continues to climb worldwide. Over the previous few decades, the industry has made great progress in reducing mortality among children under the age of five. The study aims to identify the factors associated with under-five mortality in Ethiopia.

Subject and method

Community-based Cross-sectional data came from Ethiopia's Mini Demographic and Health Survey 2019. In two stages, the 2019 EMDHS sample was stratified and selected. 8855 women of reproductive age were interviewed using a nationally representative Woman (ages 15 to 49) and 5753 children were included. Ethiopia's under-five mortality served as the study's dependent (response) or outcome variable. Binary logistic regression was used to see if there is an association between the dependent and independent variables. All variables with a p-value of less than 0.25 in the bivariate analysis were chosen for the multivariable logistic regression to compensate for putative confounders. Significant predictors were defined as factors with a p-value of less than 0.05.

Results

A total of 5753 under-5 mortality were enrolled in this study. 339 (5.9%) of under-five Mortality have been declared lifeless before reaching the age of five. There were 1328(23.1%) and 4425(76.1%) with 72 (5.42%) and 267 (6.02%) of under-five mortality occurring in urban and rural respectively. Under-five mortality in the Afar region was 2.280 times more likely Compared to Children born in Tigray Region (AOR = 2.280 95% CI = 1.137–4.568, P = .020). Under-five Mortality in Rural residences was 1.908 times more likely as Compared to Urban Residence (AOR = 1.908, 95% CI = 1.257–4.539, P = .035). Under-five mortality in Poorer index Households was 0.343 times Less likely as compared to children born in the poorest index Household (AOR = 0.343,95% CI = 0.128–0.910, P = ). Under-five mortality in the public sector was 1.763 times less likely than among children born at Home (AOR = 1.763, 95% CI = 1.252–2.482, P = .033). Under-five Mortality of second multiple births was 2.389 times more likely Compared to Single birth (AOR = 2.389, 95% CI = 1.257–4.539, P = .008).

Conclusion

This study found that the prevalence of under-five mortality, is 5.9% (59/1000) or 59 death per one thousand live children in Ethiopia. The under-five mortality rate is rapidly declining, and access to and utilization of health care is improving. Region, residence, level of education, wealth index, Place of Delivery and multiple births have all been statistically significant factors of under-five mortality in Ethiopia. The government and all stockholders should be given attention to maternal and infant health care to reduce under-five mortality.

•

The global number of under-five mortality remains high including in Ethiopia.

•

The prevalence of under-five mortality was 5.9% of live children in Ethiopia.

•

Region, residence, level of education, wealth index, Place of Delivery and multiple births have all been found to be statistically significant factors of under-five mortality in Ethiopia.

•

The under-five mortality rate is rapidly declining, and access to and utilization of health care is improving.

•

The government and all stockholders should be given attention to maternal and infant health care to reduce under-five mortality.