Factors Associated with Singleton Preterm Birth in Shire Suhul General Hospital, Northern Ethiopia, 2018

Background Preterm birth is the leading cause of neonatal mortality and significant health consequences to the newborn, families, and communities and tens of emotional and economic costs. Therefore, the aim of this study was to assess the magnitude of singleton preterm birth and associated factors in Shire Suhul General Hospital, Northern Ethiopia. Methods Institutional based cross-sectional study was conducted among systematically selected 325 postnatal mothers in Shire Suhul General Hospital, Northern Ethiopia, from January to March 2018. The data were collected through both face-to-face interview and chart review by using pretested semistructured questionnaires. SPSS version 20 was employed to enter and analyze the data. Both bivariate and multivariate logistic regression models were run to identify factors associated with singleton preterm birth at the level of P values ≤ 0.25 and <0.05 for each model, respectively. Result The magnitude of singleton preterm birth in Suhul Hospital was 16.9%. Smoking cigarette/drinking alcohol during pregnancy [AOR=3.61: CI 95%; 1.59-8.23], previous abortion [AOR=2.37: CI 95%; 1.15-4.88], hemoglobin level < 11gm/dl [AOR=2.44: CI 95%; 1.14-5.22], visible physical neonatal congenital anomaly [AOR=10.42: CI 95%; 1.66-65.23] , and history of giving low birth weight baby [AOR: 2.78 CI 95%; 1.39-5.55] were the factors statistically associated with singleton preterm birth. Conclusion The magnitude of preterm birth in this study was higher than the average prevalence of preterm birth reported in Ethiopia. Smoking cigarette/drinking alcohol during pregnancy, mothers who had previous abortion, low maternal hemoglobin level, presence of visible physical congenital anomalies of newborn baby, and history of bearing low birth weight baby were found to have statistically significant association with singleton preterm birth. Supplement of daily iron with folic acid (folic acid ideally before pregnancy) for all pregnant mothers with good adherence monitoring and giving attention to preconceptional health care service to avoid any cigarette smoking/alcohol drinking and risk detection were set as recommendations.


Introduction
The World Health Organization (WHO) defines preterm birth as a baby born too early or before 37 complete weeks of gestation from the first day of the last normal menstrual period (LNMP). Birth too soon is categorized as extremely preterm (less than 28 weeks), very preterm (28-32 weeks), and moderate to late preterm (32weeks to 36 weeks and 6 days). Preterm birth may happen spontaneously or is initiated by clinicians for different medical or nonmedical reasons and mothers deliver either single or multiple neonates. Globally, around 1 million children die each year due to complications of preterm birth and 60% of them occur in Africa and South Asia. Generally, the average rate of preterm birth in lower income countries and higher income countries is 12% and 9%, respectively [1]. In Ethiopia, 320,000 premature babies are born each year and it directly contributes to 24,400 (12.5%) deaths of under-five children [2].
Preterm birth is the major single burden for both developing and developed countries. Some of the anticipated complications following birth of preterm neonates are motor development, behavior and academic performance impairment, and even deaths within minutes [3,4]. Preterm birth often requires more resources for health care, social based rehabilitation service, special education placement, and specialized health care professionals throughout their childhood and even adulthood [5].
A high epidemic of preterm birth and related newborn deaths is one of the greatest health challenges of these days. However, two-thirds of these complications and deaths could be prevented by simple inexpensive measures that do not need high technology [22,23]. In developed countries, treatment with metronidazole and erythromycin helps to prevent premature delivery caused by bacterial vaginosis [24]. Using vaginal progesterone is the most efficacious and safe method to prevent preterm labor that leads to preterm birth [25].
However, preterm birth is a complex multifactorial process associated with diverse pathogenetic mechanisms. Most factors are usually prevented by simple strategies. Health care providers or other stakeholders who worked in this study setting need data related to common regional factors, since data limitation is there. Few studies conducted in Ethiopia are related to preterm birth. These studies were conducted by direct mothers' interview, and in the meantime, researchers have missed very important variables that need to be examined but valid data were accessible from maternal charts like maternal hemoglobin level, neonatal weight, gestational age determined by clinicians (using ultrasound or LNMP), and diagnosed medical diseases. Additionally, almost all studies included mothers who gave multiple (twin, triple, and higher) and singleton fetus indifferently. Meanwhile, multifetal pregnancies do not have an equal chance suffering for preterm birth compared to singleton birth due to physiological reasons. Therefore, answering the above issues and using it as input for the study area to conduct further studies were the aim of this study.

Study Design and Population.
Institutional based crosssectional study was conducted from January to March 2018 with 325 mothers who gave birth in Shire Suhul General Hospital at the time of data collection. The study was conducted at Shire town, the capital of northwest zone of Tigray regional state located 1087Kms far from Addis Ababa (the capital city of Ethiopia) and 357 Kms from Mekelle (the capital of Tigray regional state). Based on the reports from the zonal administrative office, the town has 67,243 total populations; among these, 36,146 are females. In Suhul General Hospital, the total annual birth report in 2017/18 was 3,048 and the first quarter (July-October) birth was 720.

Inclusion and Exclusion Criteria.
Mothers who gave birth and had known either LNMP or 1st trimester ultrasound diagnosis were included for this study.

Sample Size and Sampling
Technique. Sample size for this study was calculated by using single population proportion formula with the assumption of 95% confidence level, 25.9% prevalence of preterm birth [19], and 5% marginal error. The final sample size after considering 10% of nonresponse rate was 325. Finally, a systematic random sampling technique with sampling interval (K th ) value of two (2) was used to identify interviewed participants. The first study participant was selected randomly and continued every two intervals. are parity, gestational age of current baby in weeks, medication intake during the most recent pregnancy, smoking cigarette/drinking alcohol during the most recent pregnancy, history of hyperemesis-gravidarum in the most recent pregnancy, previous abortion, previous stillbirth, previous preterm birth, previous cesarean delivery, history of vaginal bleeding during the most recent pregnancy (antepartum hemorrhage), history of diagnosed urinary tract infection in the most recent pregnancy, hypertension disorder during the most recent pregnancy, history of hospitalization during the most recent pregnancy, previous diagnosed fibroid, maternal current serostatus, level of the most recent hemoglobin, previous malaria attack, mode of delivery of the current baby, history of ANC (antenatal care) follow-up, number of ANC follow-ups, and birth (interdelivery) interval. Fetal factors are congenital abnormalities of the current newborn, previous multiple delivery, giving congenitally defected baby, and history of giving LBW (low birth weight) baby including the most recent and sex of the current newborn.

Data Collection and Quality
Control. The data were collected by using a pretested semistructured questionnaire through both face-to-face interview and extracting (review) respective mothers' chart for necessary data. Questionnaires were prepared first in English after reviewing different works of literatures and translated into local language (Tigrigna) and then back to English by the third party who was native in Tigrigna to ensure consistency. The data collection was carried out within 24 hours before mothers going to be discharged. Four midwives who were fluent in local language were involved in data collection. Detail orientation was provided for data collectors.  Descriptive statistics were summarized into frequency and proportion. Both bivariate and multivariate analysis model were run to check the association between dependent and independent variables. All variables with P value ≤ 0.25 in bivariate analysis were taken into multivariate logistic regression model and variables having P value < 0.05 with 95% CI declared factors significant association with singleton preterm birth.

Sociodemographic Characteristics.
A total of 325 mothers participated in this study with 100% response rate. Two hundred seventy-three (84%) of the participants were of age <35 years with the mean age of 27.88 years±5 years of SD. More than three-fourths (88.6%) of the study participants were Orthodox Christian followers and majority, 306(94.2%), were married.
More than half (57.8%) of the mothers were from the urban area and 201 (61.8%) of them were the housewives in occupational status. More than three-fourths, 252 (77.5%), were primary school and less in their educational status. Family size of more than half, 189 (58.2%), of the mothers was within the range of 2-4 individuals. Almost all, 304 (93.5%), mothers have never suffered from hard workload in the most recent pregnancy and 91 (28%) participants' monthly income was within the range of 1401-2350 Ethiopian Birr ( see Table 1).

Maternal (Obstetrical and Medical)
Factors. According to this study, greater than three-fourths of the study participants gave term neonates followed by 16.9% of preterm birth (see Figure 1).

Fetal Factor.
Among newborns who were delivered during the study period, only 6 (1.8%) mothers gave congenital abnormal babies whereas 17 (5.2%) of the respondents had a previous history of giving congenital anomalies baby. Including the most recent births, 60 (18.5%) of the study participants had the history of having low birth weight baby. More than half, 183 (56.3%), neonates were female (see Table 3).

Factors Associated with Singleton Preterm Birth.
All independent variables were analyzed using binary logistic regressions and all variables with P value ≤ 0.25 were fitted to run in multivariate logistic regression. Finally, in multivariate logistic regression model, smoking cigarette/drinking alcohol during the most recent pregnancy, history of abortion, hemoglobin level, a physical congenital defect in the most recent baby, and history of bearing low birth weight baby including the most recent one were identified as independently associated variables with singleton preterm birth at P value < 0.05. Mothers who smoked cigarette/drank alcohol during the most recent pregnancy were 3.6 times more likely to have a preterm delivery than those who did not smoke/drink [AOR=3.6; 95%CI: 1.59-8.23]. Similarly, mothers who had a history of abortion were nearly 2.4 times more likely to give preterm baby than their counterparts [AOR=2.37; 95%CI: 1.15-4.88].
Mothers whose most recent hemoglobin level was <11 gm/dl were 2.4 times more risky for preterm birth as compared to those whose hemoglobin level was ≥ 11gm/dl [AOR=2.44; 95%CI: 1.14-5.22]. Finally mothers who gave neonates with physical congenital defect and who had history of delivering low birth weight baby including the most recent birth were statistically significant with singleton premature birth compared to their counterparts [AOR=10.4; 95%CI: 1.66-65.2] and [ AOR=2.78; 95%CI: 1.39-5.55], respectively ( see Table 4).

Discussions
This study revealed that the prevalence of preterm birth was 16.9%. This finding is in line with those studies conducted in India 18.01% [26], Nigeria 16.8% [27], Malawi 16.3% [16], Tanzania 14.2% [12], and Kenya 18.3% [9]. This similarity could be due to some related leveled socioeconomic status and lifestyle among participants since all are developing countries.
This finding is higher than the average rate of preterm birth in lower income countries reported by WHO 12% [28], Brazil 12.3% [29], Iran 5.1% [7], Australia 6.8% [21], Debre Markos 11.6% [30], Gondar 4.4% [14], average Ethiopian prematurity birth 10% [23], and Tigray 8.1% [31]. This disparity could be due to this study setting having low access to quality of health care this study carried being out in single hospital and small sample participants involved in this study. However, most studies conducted in the above regions were case-control, involved multiple institutions, and used a large sample size.
On the other side, the magnitude of preterm birth in this study was much lower than the study carried out in India 25.6% [32], Indonesia 35% [15], and Jimma University Specialized Hospital 25.9% [19]. This variation might be due to the fact that multiple pregnancies were not included in this study, nature of study areas, and health-seeking behavior among the study participants.
Journal of Pregnancy 5   Mothers who smoked cigarette/drank alcohol during the most recent pregnancy experienced preterm birth more than those who never smoked/drank [AOR=3.61; 95%CI: 1.59- 8.23]. This finding was consistent with that of the study conducted in China, Australia, and Jimma [19][20][21]. This could be explained by the fact that cigarette/alcohol may interfere with the nutritional process and affect the strength of the membrane negatively that leads to premature rupture of the membrane. Mothers who had previous history of abortion were also 2.4 times more likely to give preterm baby [AOR=2.37; 95%CI: 1.15-4.88] than those who never had. This is in agreement with the study conducted in Iran, Tanzania, and Jimma [12,17,19]. This finding could be explained by the fact that infection following the procedure may result in intra-amniotic infection and possibly idiopathic risk factor for preterm premature rupture of membrane and then result in preterm labor.
Mothers whose most recent hemoglobin levels <11gm/dl were nearly 2.5 times more likely to experience preterm birth compared to those ≥11gm/dl [AOR=2.44; 95%CI: 1. 14-5.22]. This figure is in line with WHO 2013 report and the studies carried out in India, Indonesia, Malawi, and Debre Markos [13,15,16,30,32]. This could be explained biologically, anemia (<11gm/dl) cause for hypoxia which can induce maternal and fetal stress, which stimulate the production of the corticotrophin-releasing hormone to lead the initiation of preterm labor.
Visible physical neonatal congenital anomalies in the most recent baby increased odds of the preterm birth occurrence compared to normal babies with [AOR=10.4; 95%CI: 1.66-65.2]. This finding was consistent with that of the study conducted in the United States of America, Palestine, and Jimma [11,18,19]. This could be because of clinicians who intended to initiate the labor artificially as soon as they detected congenital anomalies fetus who will not be compatible in life irrespective of gestational age. Mothers who had the previous history (including the most recent) of bearing neonatal weight <2500gm were 2.8fold more likely to give preterm baby compared to their counterparts. This finding was similar to the study carried out in Tanzania [12]. The possible reason related to indirect relations was because that problems with the placenta and maternal infection which could prevent oxygen and nutrients to reach the fetus result in both preterm and low birth weight baby.

Limitation of the Study
The study was conducted with a small sample size, only in one institution, and used maternal charts as additional data source which could result in overreporting and difficulty in inferring the entire society. Additionally, we failed to assess hypertension disorders during pregnancy separately as preeclampsia, eclampsia, gestational hypertension, and superimposed preeclampsia. Finally, the study design was cross-sectional, which did not show the causality of the limitations of this study.

Conclusion
The magnitude of preterm birth in this study was higher than the average prevalence of preterm birth reported in Ethiopia. Substance intake during pregnancy, previous abortion, low maternal hemoglobin level, the presence of visible physical congenital anomalies of newborn baby, and history of bearing low birth weight baby had statistically significant association with preterm birth. Supplement of daily iron with folic acid (folic acid ideally before pregnancy) for all pregnant mothers with good adherence monitoring and giving attention to preconceptional health care service to avoid cigarette/alcohol intake, and risk detection were set as recommendations.
Abbreviation AOR: Adjusted odds ratio ANC: Antenatal care COR: Crude odds ratio LNMP: Last normal menstrual period SPSS: Statistical package for social sciences WHO: World Health Organization.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
Our study received ethical approval from an institutional review board of Aksum University, College of Health Science and Comprehensive Specialized Hospital (Ref No AK/UN/R/H/CARD/mid/973/10). Suhul hospital administrative office also gave the permission and letter of approval submitted to the maternity to carry out the study.

Consent
Verbal consent was taken from the study participants after explaining the purpose of the study. Confidentiality was assured by using codes and interviewing alone with privacy. Participants were informed about the right not to participate in or withdraw from the study at any time through the interviews.