Adverse birth outcomes among women who gave births at Eastern Ethiopian Hospitals: a cross sectional study

Adverse births outcomes (ABO) such as preterm birth, low birth weight (LBW) and stillbirth are important determining of neonatal morbidity and mortality. It is the major public health problem in low resourced countries. Despite many efforts to reduce neonatal morbidity and mortality, adverse birth outcomes in hospitals has remained high. Therefore, this study aimed to assess the prevalence and factors associated with adverse birth outcomes among women who gave birth at selected public hospitals in Eastern Ethiopia. Methods Facility based cross-sectional study design was conducted in February 2017. Data were collected using a pretested and structured face to face interviewer-administered questionnaire. Binary logistic regression was used to analyze the association between the dependent and independent variables.


Results
A total of 555 women who give births were involved from four hospitals in Eastern Ethiopia. The prevalence of low birth weight, stillbirth and preterm births were 40(7.2%), 37(6.7%) and 28(5%) respectively. The overall prevalence of adverse birth outcomes was 76(13.7%).

Conclusion
In this study adverse birth outcomes were still found as public health problems. Hypertension, history of adverse birth outcome, multiple pregnancy and spontaneous vaginal delivery were signi cantly associated with ABO. Thus, developing strategies to prevent and treat complications during pregnancy and increasing maternal health utilization at ANC clinics are warranted.

Background
Adverse birth outcomes are the most health problem that leads to neonatal morbidity and mortality over the world. It includes preterm birth, low birth weight (LBW), and stillbirth (1). Worldwide, around 15 million preterm babies are born every year. Almost 35% of neonatal mortality is related to preterm birth and 99% occur in low-income countries (2,3).
In resourced countries, the rate of stillbirth is 3 to 4 per thousand births, whereas 10 folds higher in low resourced countries (4). Around 3.3 million stillbirths occurred in low resourced countries. In sub-Saharan Africa, an estimated 900,000 babies die due to stillbirths and two-thirds of all stillbirth mortality happened before the onset of labor, or antepartum stillbirths (5).
Low birth weight (LBW) also the leading cause of neonatal death. It is a major contributor to infant and under-ve mortality. Globally, greater than 20 million infants are born LBW (15.5% of all births) which 95.6% are in low resourced countries (6).
Early and late morbid conditions like coronary heart disease, non-insulin dependent diabetes, childhood hypertension, behavioral disorders, impaired cognitive function, psychological disorders and long-term nancial burden are risk factors associated with LBW(7).
In Ethiopia, a high rate of neonatal mortality is reported and preterm birth is a major and direct cause of neonatal mortality. According to the Ethiopian demographic health survey (EDHS-2016), a ve years pro le of infant and child mortality indicated that the neonatal mortality rate was declined from 49 deaths per 1,000 live births in 2000 to 29 deaths per 1,000 births in 2016 (8).
Despite numerous strategies and interventions, prevention of adverse birth outcomes have remained largely unaddressed. In general, epidemiological data on the magnitude and risk factors of adverse birth outcomes are limited in the study area. Therefore, this study aimed to determine the prevalence and associated factors of adverse birth outcomes at selected public hospitals in Eastern Ethiopia.

Methods
A facility based cross-sectional study design was employed in February 2017 at Eastern Ethiopia Public Hospitals. Four Hospitals: Hiwot Fana Specialized University Hospital, Chiro Hospital, Karamara Hospital and Dilchora Hospital were selected randomly from Harari regional state, Oromiya regional state, Somali regional state, and Dire Dawa city Administrative respectively (9)(10)(11)(12). All women who gave birth in the selected study hospitals during data collection period were included in this study. Women with serious illness who were unable to respond were excluded Data collection tools and procedure: A pretested and standardized face-to-face interviewer-administered questionnaire was used to collect data. The questionnaire contains socio-demographic characteristics, maternal and obstetric history, and history of chronic diseases. Data collection tool was prepared in English and translated to local languages (Amharic, Afan Oromo, and Somali) for administration and then translated back to English to maintain consistency. Four Diploma Midwives and two Bsc Midwives who were uently speaking local language were recruited for data collection and supervision respectively.
The current pregnancy outcomes were dichotomized into favorable birth outcomes and adverse birth outcomes. Favorable birth outcome considered if the delivery resulted in a live term baby with normal birth weight. Adverse Birth Outcome refers delivery of preterm birth (between 28 and 37 weeks gestation calculated from LNMP or early ultrasound reading), low birth weight(less than 2500 g after delivery) or stillbirth (any fetus born at 28 weeks gestational age or more with no heartbeat or respiratory effort after viability) (6,13,14).
Data processing and analysis: The collected data were entered into the computer by using Epi info-version 7. Data cleaning and analysis was done using SPSS-version 20. Bivariate and multivariate logistic regression analysis was used to analyze the association between the dependent and independent variables. Hosmer-Lemeshow's and Omnibus tests were done to test for model tness. Bivariate logistic regression was used to identify predictors of adverse birth outcomes. All variables with p-value ≤ 0.2 were taken into the multivariable model to control for all possible confounders. Finally, the results of multivariable logistic regression analysis were presented in crude and adjusted odds ratios with 95% con dence intervals. The level of statistical signi cance was declared at a p-value of less than 0.05.

Ethical considerations:
Ethical clearance for the study was obtained from the Institute of Health Research and Ethical Review Committee of Haramaya University College of Medical and Health Sciences, regarding written and ngerprint consent, the privacy of participants, risk and bene t analysis of the study were approved.
Informed voluntary written and ngerprint consent was obtained from study participants after explaining the purpose and procedure of the study.

Results
A total of 555 study participants were participated in this study. The mean age was 26.6 (± standard deviation = 4.62) years. The majority were Muslim (65.8%), and (63.6%) of study subjects were Oromo ethnics. Most (94.7%) of study participants were married, and more than half (51.4%) were housewives. One third (33.3%) of the participants had no formal education. The mean age at rst marriage was 20.8 (± standard deviation = 3.5) years (Table 1). Maternal and obstetric history of study participants: The majority of the study participants (65.8%) were multi-parous and 357(64.3%) of study participants had taken iron supplementation during their pregnancy follow up. Greater than three-fourth (79.5%) of study participants had ANC follow up at least one visit (Table 2).
Follow up of ANC was assessed: majority 121(27.4%) of pregnant women had ANC four visits, on the other hand, around 61(13.8%) of pregnant mothers had merely one follow up before the onset of delivery.
The frequencies of ANC follow up from second to fourth visits were similar, and only 29(6.6%) pregnant women had ANC four-plus visits (Fig. 1).
Fifth-seventh 399(71.9%) of pregnant women who participated in the study had a tetanus toxoid (TT) vaccine ( Table 2). The TT vaccines protect the women from tetanus during birth if the pregnant mothers take TT2 and above. Majority 346(86.7%) of study participants had taken TT2 and above while 53(13.3%) had only one dose before their delivery from 399 participants (Fig. 2).  Prevalence of adverse birth outcome: The prevalence of adverse birth outcomes was 40(7.2%) of low birth weight, 37(6.7%) of stillbirth, and 28(5%) of preterm birth. Among all deliveries 76(13.7%) had resulted in the adverse birth outcomes and 479(86.3%) resulted in favorable birth outcomes (Fig. 3).
Factors associated with adverse birth outcome: The ndings from bivariate logistic regression revealed that residence, maternal education, chronic disease(hypertension), history of adverse birth outcome, multiple pregnancy, pregnancy-induced hypertension, antepartum hemorrhage and mode of delivery (spontaneous vaginal delivery) were the independent predictors of adverse birth outcomes. However, in the multivariate logistic regression analysis, hypertension, history of adverse birth outcomes, and spontaneous vaginal delivery remained signi cantly (Table 3).
Study participants who had diagnosed hypertension were seven times more likely to have adverse birth outcomes than normo-tensive women [AOR = 7.25; 95%, CI: (1.71, 30.64)]. The odds of adverse birth outcome was about twelve times higher among women with a previous history of adverse birth outcome than those women who did not have previous history of adverse birth outcomes [AOR = 12.12; 95%, CI: (6.5, 22.6)]. The odds of adverse birth outcomes were about seven times higher among women with multiple pregnancies than those women with singleton pregnancy [AOR = 6.94; 95%, CI: (2.74, 17.53)]. The odds of adverse birth outcomes were less among women who gave birth through spontaneous vaginal delivery than those women who delivered by cesearean section [AOR = 0.11; 95%, CI: (0.44, 0.16)] (Table  3).

Discussion
In this study, we assessed the prevalence and associated factors of adverse birth outcomes (stillbirth, preterm birth and low birth weight) among women who gave birth at selected public hospitals in Eastern Ethiopia. The overall prevalence of adverse birth outcomes among the study participants was 13.7%. This nding is relatively similar with studies nding from Kembata Tembaro zone(13.9%) (15) and Hawassa (18.3%) (16), but lower than that of Gondar (23%)(17), Hossana town (24.5%) (18) and Wollo zone (31.8%) (19). This variation may be due to the difference in the study setting, socio-cultural status, maternal and newborn health care services and various interventions undertaken between these study times.
However, lower than the study done in Wollo (19.5%) (19). The possible explanation for this discrepancy may be attributable to socio-economic variations and the difference in the quality of maternal health service. Also, the variation of nutritional and cultural practices may contribute to the observed differences.
In this study the prevalence of preterm birth was 5%. This nding is consistent with that of Hawassa (3.6%) (16) and Butajira (6.4%) (18) but lower than that of Wollo (13.2%) (19) and Gondar (14.3%) (17). The difference may be due to variation in populations and study setting and quality of maternal healthcare service.
The ndings from the multiple logistic regression analysis revealed that hypertension, history of adverse birth outcomes, multiple pregnancy, and spontaneous vaginal delivery were signi cantly associated with adverse birth outcomes. This nding was almost found to be a universal fact and has been revealed in many studies (16,17,20).
This study showed that women who had diagnosed hypertension were statistically signi cant with adverse birth outcomes. This nding is in line with studies nding from china, Tanzania and Ethiopia (17,(20)(21)(22). It can be justi ed by uteroplacental hypo-perfusion secondary to vasoconstriction which results in diminished blood supply to the fetus causing adverse fetal/birth outcome.
In our nding women with a previous history of adverse birth outcomes were signi cantly associated with adverse birth outcomes. This is in line with the report from other studies from china, Tanzania and Ethiopia (16,17,21,22). This might be due to most bad obstetrics histories are recurrent and more likely to result in adverse birth outcome in subsequent pregnancies.
The study nding showed that women with multiple pregnancies were signi cantly associated with adverse birth outcomes. This nding is similar to other studies nding from Ethiopia (20,23). This might be due to chorionicity and amniocity of placentation which is associated with fetal mortality and morbidity in monochorionic and mono amniotic twins. There was also a complication unique to twin pregnancy which is twin transfusion syndrome, discordant twin, conjoined twin, vanishing effect which increases the risk of adverse birth outcomes. The risk of anemia and pregnancy-related complications also increase in multiple pregnancies hence more likely to result in adverse birth outcomes.
The study also revealed that the odds of adverse birth outcome is less by 11% among women who gave birth through spontaneous vaginal delivery than those women who delivered by cesearean section. This nding is in contrast with other studies ndings from china and WHO Multi-country Surveys (24,25). The difference may be due to study setting as those two studies were large and were conducted at better health care set up were ceaseran delivery performed on elective basis without indication than the current study. Finally, as limitations the ndings of this study were based on self-report, as it was not possible to validate claims made by respondents in the course of questionnaire administration.

Conclusion
Adverse birth outcome in the study area was still a publicly signi cant problem. Hypertension, history of adverse birth outcomes, multiple pregnancies, and spontaneous vaginal delivery were signi cantly associated with adverse birth outcomes. Hence, increasing access to maternal health utilization and developing strategies to prevent and treat complications during pregnancy is recommended. submitted to all concerned bodies in the study area to secure their co-operation in facilitating the study.

List Of Abbreviations
All the participants were asked for voluntary informed written and ngerprint consent for participation before interview initiation.

Consent for publication N/A
Availability of data and material The data for this study can be made available from the corresponding author when there is a reasonable need. Figure 1 Number of ANC follow up or visit among study participants at selected public hospitals in Eastern Ethiopia, 2017(n=441).