Time to first birth and its determinants among married female youths in Ethiopia, 2020: survival analysis based on EDHS 2016

Introduction: The first birth is the most significant events in a woman’s life that indicates the beginning of undertaking the intensive responsibilities of motherhood and childcare. Age at first birth has health, economic and social consequences and implications. But little has been known on the time to first birth and its determinants in Ethiopia. Therefore, this research is planned to address this issue. Objectives of the study: To assess the time to first birth and its determinants among married female youths in Ethiopia, 2020. Methods: The data was accessed freely through (https://www.dhsprogram.com). Survival analysis of time to first birth was done based on EDHS 2016 data among 2597 weighted study subjects. The data was extracted using STATA version 14.0. Kaplan Meier’s survival and Log rank test were used to compare survival experiences of respondents using categorical variables. Proportional hazard assumption was checked and was not violated. Cox proportional hazard model was applied, hazard ratio with 95% CI was computed and variables with p value < 0.05 in the multivariable analysis were taken as significant determinants. Results: Overall median survival time was 18 years (IQR = 17–20). The significant determinants of time to first birth are place of residence (being rural (AHR = 1.49, 95% CI 1.13, 1.97),Religion (being Muslim [AHR = 1.57, 95% CI 1.22, 2.02),being protestant (AHR = 1.73, 95% CI 1.34, 2.24)], age at first sex [first sex < 15 years (AHR = 1.68, 95% CI 1.23, 2.29)] and first sex between 15 and 17 years (AHR = 1.54, 95% CI 1.29, 1.85), age at first marriage (marriage < 15 years (AHR = 6.52, 95% CI 4.91, 8.64), marriage between 15 and 17 (AHR = 2.63, 95% CI 2.20, 3.14), unmet need for family planning (AHR = 1.23, 95% CI 1.00, 1.52) Conclusion: In this study, the median age at first birth was 18 years. This show, about 50% of study participants give birth for the first time before their 18th birth day. This age is the ideal age for schooling and to do other personal development activities. Therefore giving birth before 18 year will limit female youths from attending school and performing personal development activities in addition to health and demographic consequences of early child bearing. The study was conducted in Ethiopia using the 2016 Ethiopian demographic and health survey data. In Ethiopia according to the EDHS 2016 report, the median age at first marriage and first sexual intercourse were 17.1 years and 16.6 years respectively, the maternal mortality ratio was 412 per 100,000 live births, skilled delivery coverage was 28%, the contraceptive prevalence among married women and a fertility rate was 36% and 4.46 respectively and Sexually active unmarried women accounts 58% [12]. The principal investigator requested Central Statistics Agency (CSA) to get access the data. The data was accessed freely through (https://www.dhsprogram.com). The data extraction period was from 1st June to 1st July. Weighted sample was 2597. About 2597 participants were involved after weighting. The dependent variable in this study was the time (age in years) at first birth. Whereas the independent variables were Socio-demographic and economic factors (educational status, employment status, residence, religion and economic status), reproductive factors (number of miscarriages and stillbirths before first birth, use of contraceptives, age at first sex and age at first marriage), behavioral factors (cigarette smoking, chat chewing, alcohol drinking), media related factors (media exposure and hearing family planning messages on mass medias) and husband related factors (husband education, husband occupation, husbands desire for child and husband’s behavioral factors like alcohol drinking, chat chewing and cigarettes smoking). Event: giving first birth. Censored: Not giving first birth. Survival time: The time is taken in years (age) from her birth to give her first birth. A total of 2597 respondents were included in this study. All sampling procedures are available at the EDHS reports of each year. But to highlight the sampling procedures and step of data extraction we use the following diagram (Fig. 1). Schematic representation of the sampling procedures in the study of time to first birth and its determinants among married youth females in Ethiopia, 2016. The final sample size after being weighted by sampling weight was 2597. N.B EAs Enumeration Areas Survival analysis was used to identify the determinants of age at first birth. All procedures of Ethiopian Demographic and health survey were performed based on the international ethical guideline of the DHS program [12] and the data processing and analysis in this document were done based on a step-wise guide to perform survival analysis [30]. The data for this study were extracted from EDHS 2016 individual (women) record folder using STATAversion14software. The extracted data was recorded, cleaned using frequency; listing and sorting to identify any missed values. Kaplan Meier's survival was used to estimate the time of first birth and used to compare survival experiences among categories across time. A log-rank test was used to compare survival experiences between two or more groups and to show the statistical significance of the observed difference in the Kaplan Meier's survival plot. The PH assumption was checked both graphically and using global test. A global test based on Schoenfeld residuals found that all of the covariates and the full model satisfied the proportional hazard assumption (Chi square = 16.35 and p value = 0.0900). As stated above the data source for this study was the EDHS data which has hierarchical nature (that means youths in the same cluster may have similar characteristics than youths in another cluster). As a result, we have checked the presence or absence of between cluster variability/ heterogeneity by fitting frailty model (random effect Cox regression model) and the theta value at the null model was insignificant (p value > 0.05). It indicates that there is no unobserved heterogeneity between clusters. Multicollinearity was also checked and the maximum variance inflation factor (VIF) was 2.62 for age at first marriage and the mean VIF was 1.67. This shows that no multicollinearity between covariates. Variables which were associated with the outcome variable in the bivariable analysis at p value < 0.25 were entered into a multi-variable analysis. The cutoff point for the significant association was p value < 0.05 at 95% confidence interval. Survival analysis is analysis of history of events which uses statistical procedures to analyze time duration, until one or more events of interest happen. Time-to-event (TTE) data is unique because the outcome of interest is not only whether or not an event occurred, but also when that event occurred. Logistic and linear regression is not suited to be able to include both the event and time aspects as the outcome in the model. These regression methods also are not equipped to handle censoring, a special type of missing data that occurs in time-to-event analyses when subjects do not experience the event of interest during the follow-up time [29].