Spatial patterns and associated factors’ of Early Marriage among Reproductive age women in Ethiopia: a Secondary Analysis of Ethiopian Demographic and Health Survey 2016

Background : Besides, the presence of national law, the country has to set up its own mid-term and long term goals to bring about a significant reduction in child marriages in Ethiopia. To achieve this, determining the spatial pattern of early marriage and factors associated is important for government, other concerned bodies, program implementers and policy developers to end up early childhood marriage. Thus, the aim of this study was to assess the spatial patterns and associated factors of Early marriage among reproductive-age women in Ethiopia. Methods : This study analyzed retrospectively a cross-sectional data on a weighted sample of 11,646 women aged 15-49 years after requesting from Ethiopian Demographic and Health Survey 2016 via the link www.measuredhs.com. ArcGIS version 10.7 software was used to visualize spatial distribution for Early marriage. The Bernoulli model was applied using Kilduff SaTScan version 9.6 software to identify significant purely spatial clusters for Early marriage in Ethiopia multiple logistic regression analysis was used to identify factors associated with early marriage. Finally, variables with a p-value<0.05 were considered as statistically significant. Results : In this analysis, about 62.8% (95%CI: 61.9, 63.74%) of the study participants were married before they reached 18 years. The overall median age at first marriage was 17.1 with IQR 5 years. The high clustering of early marriage was located in Amhara, Afar, and Gambella Regions. In spatial Scan statistics 87 clusters (RR = 1.28, P-value < 0.001) significant primary clusters were identified. The associated factors of early marriage were lesser among women’s attending primary (AOR=0.60; 95%CI: 0.51, 0.71), secondary (AOR=0.19; 95%CI: 0.13, 0.26) and tertiary education (AOR=0.11; 95%CI: 0.07, 0.18). Similarly, women found in Addis Ababa were at a lesser risk of early marriage compared to other regions of the country. Conclusion : 18

Early marriage is defined as the marriage of a girl <18 years of age and is a common phenomenon worldwide(1). According to the United Nations Children's Fund (UNICEF), Each year, 12 million girls are married before the age of 18 years (2). The problem is highly prevalent in Asia (45%) followed by sub-Saharan Africa (39%), Latin America (23%) and 18% in the Middle East and North Africa (3). As an illustration, in Africa, the prevalence of early marriage was 31.4% in Zambia (4). Similarly, in Ethiopia, the percentage of women marrying before age 18 has declined slightly since 2011 from 63% to 58%.
During the same period, the median age at first marriage among women age 25-49 has increased from 16.5 years to 17.1 years (5).
Childbearing below the age of 18 years is associated with a higher rate of mortality, eclampsia, postpartum hemorrhage, HIV infection, malaria, and obstructed labor (6,7). In addition, early marriage is associated with lower levels of schooling for girls, higher intimate partner violence and poor maternal and child nutrition status (6). The probability of being stunted and wasting is higher among children born from early married women. The consequence of early marriage is not limited to the mother and her child, it has also social, economic, and political implications (8).
Despite the presence of national laws in Ethiopia, a marriage of girls <18 years of age is common and it affects a number of girls (9). The problem may worsen when it exists with a high prevalence of HIV and other sexually transmitted diseases, malnutrition, cervical cancer, and others. The governments of Ethiopia have adopted strategies to end the practice and investments are being made to that effect, including by promoting girls' education and sexual and reproductive health and rights. But ending child marriage requires a multifaceted approach focused on the girls, their families, the community, and the government.
In Ethiopia, several studies identified that education, harmful tradition practice, income, family size, media exposure and culture of the community were the significant factors associated with early marriage (9)(10)(11)(12)(13)(14)(15)(16). So far different studies in Ethiopia done to identify the factors associated with early marriage. The spatial pattern of early marriage was not done before. Identifying the spatial pattern of early marriage in Ethiopia can help health planners and policymakers to develop target interventions to decrease early marriage The the reseach hypothesis of this study is that is there any relationship beteewn outcome variable (early marriage) and the indeprent variables include age, religion, respondents' highest education attainment, educational status of husbands/parents, occupational status of respondents, occupational status of parents, media exposure, and household wealth status, residence, and region. The other hypothesis is that where early marriage was randomly distributed across the county. Therefore, besides the presence of national law, the country has to set up its own mid-term and long term goals to bring about a significant reduction in child marriages in Ethiopia. To achieve this, showing the spatial pattern and its factors associated are important for government, other concerned bodies, program implementers and policy developers to end early childhood marriage. Thus, the aim of this study was to assess the spatial patterns and associated factors of Early marriage among reproductive-age women in Ethiopia.

Study area, data source, sample
The study was based on the Ethiopian demographic and health survey (EDHS) 2016 data set. Ethiopia is situated in the Horn of Africa and has 9 Regional states (Afar, Amhara, Benishangul-Gumuz, Gambela, Harari, Oromia, Somali, Southern Nations, Nationalities, and People's Region (SNNP) and Tigray) and two Administrative Cities (Addis Ababa and Dire Dawa).
Approval letter for the use of this data was gained from the Measure DHS and the data set was downloaded from the Measure DHS website; www.meauredhs.com. The survey covered all the nine regions and the two city administrations of Ethiopia and participants were selected through a stratified two-stage cluster sampling technique. The full details of the methods and procedures used for the collection of the EDHS data have been published elsewhere (5).The survey collected information from a nationally representative sample of 16,683 women aged 15-49 years. Finally, 11,646 eligible women were included in this study which was nested within 643 clusters across the country.

Source and study population
The source population was all reproductive-age women within five years before the survey in Ethiopia and all women whose age between 15-49 years in the enumeration areas within five years before the survey was the study population. A total of 18008 households were selected and 16,650 were successfully interviewed. A total of 11,646 women who had married five years preceding the survey were included in this analysis (Figure 1).

Variables of the study
The outcome variable for this study was age at first marriage (binary) either below 18 or 18 and above. The variables that may influence early marriage include age, religion, respondents' highest education attainment, educational status of husbands/parents, occupational status of respondents, occupational status of parents, media exposure, and household wealth status, residence, and region(9-16).

Data collection procedure, tools, and quality control
The data was obtained from Individual Records (IR) file EDHS 2016 survey year at www.dhsprogram.com website. The web provided the data only for authorized users. Data also contained longitude and latitude coordinates. Ethiopian Demographic and Health Survey data were collected by two-stage stratified sampling. Each region of the country was stratified into urban and rural areas. The EDHS 2016 was used as a structured and pre-tested questionnaire for data collection The 2016 EDHS data collectors used tablet computers to record responses during the interview. The tablet was equipped with Bluetooth technology to enable remote electronic transfer of files for this study the detail is found at (5).

Spatial autocorrelation and hot spot analysis:
We used Arc GIS 10.7 software for spatial autocorrelation and detection of hot spot areas analysis. Spatial autocorrelation (Global Moran's I) statistic measure was used to assess whether an early marriage was dispersed, clustered, or randomly distributed in Ethiopia. Moran's I values close to −1 indicates early marriage dispersed, close to +1 indicates clustered, and if Moran's I value zero indicates randomly distributed (17). Hot Spot Analysis (Getis-Ord Gi* statistic) of the z-scores and significant p-values tells the features with either hot spot or cold spot values for the clusters spatially.

Spatial interpolation:
The spatial interpolation technique is used to predict early marriage for unsampled areas based on sampled EAs. For the prediction of unsampled EAs, we used deterministic and geostatistical Ordinary Kriging spatial interpolation technique using ArcGIS 10.7 software.

Spatial scan statistics:
We employed Bernoulli based model spatial scan statistics to determine the geographical locations of statistically significant clusters for early marriage using Kuldorff's SaTScan version 9.6 software (18).
The scanning window that moves across the study area in which early marriage was taken as cases and those women who married after age 18 and above as controls to fit the Bernoulli model. The default maximum spatial cluster size of < 50% of the population was used as an upper limit, allowing both small and large clusters to be detected, and ignored clusters that contained more than the maximum limit with the circular shape of the window. Most likely clusters were identified using pvalues and likelihood ratio tests on the basis of the 999 Monte Carlo replications.

Statistical analysis
STATA version 14 was used for data analysis. Then data cleaning was carried out. Both descriptive and analytical studies were done. Both bi-variable and multiple logistic regression analysis was performed to determine the existing association. Initially, bivariate analysis was performed and variables with a p-value of 0.2 and below were used for further analysis in the multivariable logistic regression model. At the same time, Crude Odds Ratio (COR) and Adjusted Odds Ratio (AOR) with their corresponding confidence interval (CI) also determined for the bivariate and multivariate logistic analysis, respectively. Finally, a p-value of less than 0.05 level of significance was used to declare the significance of association in the multi-variable model.

Results
In this analysis, about 62.8% (95%CI: 61.9, 63.74%) of the study participants were married before they reached 18 years. The overall median age at first marriage was 17.1 years with IQR 5. The majority, 88.95%), of the respondents were rural areas. More than half (60.61%) of respondents had no formal education. Near to three-fourth, (77.9%), of the respondents had no exposure to mass media. (Table1).

Spatial Autocorrelation
The spatial distribution of Early marriage in Ethiopia was non-random in the EDHS 2016 dataset. The global Moran's I value was 0.354 (P-value < 0.001) and Z-score 21.6 in the 2016 Ethiopian Demographic and health survey (Figure 2).

Ethiopia.
To determine spatial clustering for early marriage, global spatial statistics were estimated using Moran's I value. As shown in the figure below a statistically significant z-scores indicate at 151.3 Km distances where spatial processes promoting clustering are most pronounced. The incremental spatial Autocorrelation indicates that a total of 10 distance bands were detected with a beginning distance of 121,813 meters. (Figure 3)

Hot spot (Getis-Ord Gi) analysis:
As shown in the figure below, the red color indicates the more intense clustering of high (hot spot) proportion early marriage preceding the survey period. A high proportion of early marriage was clustered at the Amhara, Afar, and Gambella region of Ethiopia. Whereas, Amhara, SNNPR and Addis Ababa regions of Ethiopia were less risk area. (Figure 4)

Spatial Sat Scan analysis of Early marriage among women across regions of Ethiopia, 2016
Most likely (primary clusters) and secondary clusters of early marriage were identified. A total of 163 (87 primary and 76 secondary) significant clusters were identified. The primary clusters' spatial window was located in the Amhara Tigray and Benishangul regions, which was centered at 11.66 N, 37.31 E with a 254.88 km radius, and Log-Likelihood ratio (LLR) of 126.18, at p < 0.001. It showed that women within the spatial window had 1.28 times higher risk of early marriage than women outside the window. The secondary clusters' spatial window was typically located in the Somali and Oromia regions. Which was centered at 6.30 N, 41.25E with 340.06km radius, and LLR of 18.95 at pvalue <0.001 It showed that women within the spatial window had a 1.11 times higher risk of early marriage than women outside the window (Figure 5 and Table 2).

Interpolation of Early Marriage in Ethiopia
The predicted early marriage over the area increases from green to red-colored areas. The red color indicates high-risk areas of predicted early marriage and the green color indicates the predicted lowrisk areas of early marriage. The Amhara, Afar, Gambela and some parts of the Somali region, were predicted high-risk areas of early marriage. Continuous images produced by interpolating (Kriging interpolation method) early marriage among women (Figure 6) Factors associated with early marriage.
After adjusting for different confounding variables, age group, women education level and region were significantly associated with early marriage in Ethiopia. As age group increases the odds of early  Table 3).

Discussion
This study has strengths of having large dataset include thee EDHS survey and were nationally representative. Multiple logistic regression analysis was used to reduce confounding among explanatory variables. The spatial analysis was also used for identifying hotspot areas, most likely clusters and the prediction was performed to predict unsampled/unmeasured areas in the country.
Despite Ethiopia has instituted laws inhibiting marriage under 18 years and early female marriage is associated with a number of poor social and physical outcomes for young women and their offspring (19) child marriage is a norm in the country. Similarly, more than 62% of the study participants in this study were married before they reached the age of 18 years. This finding is higher than the findings from Zambia, 31.4% (20), and Ghana, 29.9% (21). This might be explained by the disparity in educational, socioeconomic and cultural differences between the study settings. The problem needs a comprehensive approach, including their families, the community, the government, and religious leaders, to reduce child marriage, teenage childbearing, and its negative consequences.
The spatial distribution of early marriage in Ethiopia was non-random. The spatial distribution of early marriage was high in Amhara, Afar, Gambella regional states of Ethiopia. The 2016 EDHS most likely significant cluster located at Amhara and Tigray and Secondary clusters were located in Somali and Oromia regional states of Ethiopia. The possible geographical variation of Early marriage Ethiopia might be socio-demographic factors, cultural behaviors.
Early marriage is often common among poor and less educated communities. A similar finding is found in this study. Women who attended primary education and above were less likely to marry before 18 years compared to their counterparts. Because educated women have a chance to determine their first age of marriage and are more likely to have a say in decision-making regarding the size of their families and the spacing of their children. In addition, educated women are also likely to be more informed and knowledgeable about contraception and the healthcare needs of their children (22,23).
In this study, women were more likely to early marriage in other regions of the country compared to Adds Ababa. It is true that poverty is one of the most powerful drivers of the harmful practices and poor families believed they will be more financially secure once their daughters are married off and out of their responsibility. In addition, poor families want to reduce the number of children to feed, clothe and educate and families may agree to child marriage because of community pressures and norms (23).

Strength and limitation of the study
This study has strengths of having large dataset include thee EDHS survey and were nationally representative. Multiple logistic regression analysis was used to reduce confounding among explanatory variables. The spatial analysis was also used for identifying hotspot areas, most likely clusters and the prediction was performed to predict unsampled/unmeasured areas in the country.
However, the limitation of this study is the cross-sectional nature of the study design may affect causality.

Conclusion
Early marriage was high in Ethiopia. High-risk area of early marriage was located in Amhara, Afar, and Gambella and special attention should be given for identified risk areas. Therefore, providing educational opportunities to young girls was important in addition to inhibiting the marriage of girls

Availability of data and materials
The datasets generated during the study are publicly available from the Measure DHS Website www.measuredhs.com Spatial autocorrelation of Early marriage in Ethiopia among the reproductive age group in EDHS 2016. The spatial autocorrelation of early marriage among reproductive age group women in Ethiopia by a function of distance. Interpolation of high-risk fertility behavior among reproductive-age women in Ethiopia, 2016