Prevalence of Neonatal Mortality and Its Associated Factor among Neonates Who Admitted In Neonatal Intensive Care Unit At Debre Tabor General Hospital, South Gondar, Amhara, Ethiopia

Background Neonatal mortality is the death of newborn babies from the time of birth to 28 completed days of life which are the most vulnerable time for a child's survival. About one million of them passed away on their rst day of life, and more than two thirds (38%) of the deaths were in sub-Saharan Africa where Ethiopia is one of the countries with the highest neonatal mortality in the world which accounts for 29 deaths per 1,000 live births. Objective To assess the prevalence and associated factors of neonatal mortality among neonates admitted in Debre Tabor General Hospital in South Gondar, Ethiopia. Methods Institutional based retrospective cross-sectional study design was conducted from November 1, 2018, up to January 30, 2019, in Debre Tabor General Hospital. A Structured interviewer-administered pre-tested questionnaire was used to collect data. The collected data were entered into Epi data version 4.2 and then exported into SPSS window version 24. Bivariate and multivariate analysis was undertaken and information was presented by using simple frequency tables, graphs, and pie charts.


Introduction
Neonatal morbidity and mortality are the major global public health challenges, speci cally for the rst 28 days of life. Globally, each year over 4 million neonates died within 28 days of birth among 130 million births. Every minute, twenty under-ve children die, leading to 8 million deaths before they reach their fth birthday due to the conditions which could be either avoided or treated. Neonatal mortality accounts for two-thirds of deaths of infants, and nearly two-fths of all deaths in under-ve children. Many babies die nameless and unrecorded, indicating the perceived inevitability of their deaths. Most neonatal deaths (99%) occur in low and middle-income countries, where about half of the deaths occur at home [2].
Ethiopia, one of the countries with the highest neonatal mortality in the world, is responsible for 29 deaths per 1,000 live births which was over 9 times more than that of highly developed countries, where the rate is 29 per 1,000 live births [1].
The health of future societies depends on the health of the children of today and their mothers. The neonatal period is considered as the highest risk period. Childhood mortality is often used as a broad indicator of social development or a speci c indicator of the health conditions of a country. However, child health programs were given low attention, especially neonatal health [5].
Every year an estimated 4 million babies die in the rst 4 weeks of life (the neonatal period). Threequarters of neonatal death happens in the rst week, the highest risk of death is on the rst day of life.
The highest number of neonatal death was in South Asian countries and the highest rate is generally in sub-Saharan Africa. Preventing death in newborn babies has not been a focus of child survival or safe motherhood programs. While we neglect these challenges, 450 newborn die every hour, mainly from the preventable cause which is unjusti able in the 21 century [2].
Neonatal mortality accounts for 44% of under-ve mortality in 2014. Average death during the neonatal period is 30 times higher compared to the rest of under-ve children.
Ten countries account for 67% of neonatal mortality globally by which Ethiopia accounts for 4% of global neonatal mortality [6].
Neonatal mortality has three main causes in low and middle-income countries. The complication of preterm, asphyxia, and neonatal infection together contribute to 85% of newborn death.
In 2013, 35% of the global neonatal deaths were caused by complications of preterm birth, 24% by intrapartum related complications, and 25% by infection. The rest of the death is caused by congenital malformations [7,8].
In Ethiopia, the rate of under-ve mortality (U5MR) decreased by 60% from the year 2000 to 2016 which implies from 123 to 67 per 1000 of live births. However, the neonatal mortality rate (NMR) decreased only by 40% from 49/1000 live births to 29 per 1000 live births from 2000 to 2016. Because of this, the share of neonatal mortality in under-ve mortality has been increased from 29.5% to 43% [3].
Neonatal health problems are usually seen as similar to an older child's health problem. But, the causes of neonatal mortality and intervention to improve neonatal health are different from that of other underve children. Neonatal survival is the most important indicator of improved health care during childbirth.
There is a gap in integrating the need for care of neonates in the NICU, the neonatal health problems, and health programs for maternal and child health. As a result, there is still slow progress to decrease neonatal mortality in the majority part of the country by which they contribute to the national burden of neonatal mortality in Ethiopia.
Even though there were some endeavors in identifying major causes of neonatal death, in Ethiopia, studies that identify the prevalence and associated factors of neonatal mortality were limited speci cally for hospital admitted neonates. Having data on the prevalence of the problem is crucial to take prioritized actions. Therefore, this study has done to identify the prevalence of neonatal mortality and associated factors among NICU admitted neonates.

Study Design and period
An Institutional based cross-sectional study was conducted from November 1, 2018, to January 30, 2019, in Debre Tabor General Hospital in NICU among admitted neonates.

Study setting
Debre Tabor General Hospital is found in the South Gondar zone in the Amhara region of Ethiopia. It was o cially commenced its function in 1917 and currently, it delivers the health care services for more than 2.3 million populations through medical, surgical, gynecological, pediatrics, NICU, ophthalmological wards, and 14 OPD with a total of 182 beds and 444 staffs.
Annually, nearly more than 1250 neonates were admitted to the neonatal intensive care unit with different health problems. It has 22 beds, 3 pediatricians, and 20 nurses in which 2 are neonatal nurses.

Study population
All neonates who were admitted at NICU in Debre Tabor General Hospital from November 1, 2018, up to January 30, 2019.

Sample size estimation
The sample size was calculated using single population proportion formula and by taking into the following consideration: prevalence (P) of neonatal mortality 13.29% [17] con dence level (CL) 95%, the margin of error (d) 5%, and by adding 10% for non-response rate. The nal sample size came up to 195.
However, to make a statistical analysis of logistic binary logistic regression, a 50% proportion (p) was used to get the nal sample size to be 422.

Data collection and procedure
Before data collection, both the data collectors and supervisors have trained by the principal investigator for one day. Secondary data was collected from Neonatal ICU treatment registers by using a structured questionnaire and checklist designed in the way that it could collect pieces of information about all the relevant variables. Then it was pre-tested with individuals equivalent to 5% of the calculated sample size among Debre tabor General hospital. Data were collected by two Bsc nurses and one Msc nurse for the supervisor. The data has collected among admitted neonates on a speci ed period from NICU. A simple random sampling method was used to select neonates admitted in NICU from medical records. This was done by sampling frame 1up to N [1; 1000] admitted neonates in NICU, then choose 422 samples were chosen every 2 intervals of medical record systematically, so it was taken the sample from a patient chart and register which have full information.
Data quality control Data quality was assured by proper designing of the questionnaire in Amharic the local language to prevent misinterpretation, apply pre-test on 5% of the sample in Debre Tabor General Hospital. The data have also coded, entered, cleaned, and made close supervision during data collection by the principal investigator.

Data analysis
Then data have coded, entered, and cleaned using Epi-data version 4.2 software and nally exported into SPSS version 24 for analysis. Bivariate analysis, crude odds ratio with 95% CI, was used to see the association between each Independent variable and the outcome variable by using binary logistic regression. Independent Variables with a p-value of ≤ 0.05 were included in the multivariable analysis to control Confounding factors. Adjusted odds ratios with 95% CI were estimated to identify the prevalence and associated factors of neonatal mortality using multivariable logistic regression analysis. from mothers with no formal education and 41 (9.7%) were from mothers with completed primary educational level. The majority of neonates were from housewife mothers and the least was from other occupations which account for 223 (52.8%) and 80 (19 %) respectively ( Table 1).

Prevalence of neonatal mortality
The prevalence of neonatal mortality was found to be 12.3% ( gure 1).
Most neonates died at early neonatal age and late neonatal age which accounts for 45 (87%) and 7 (13) respectively. Sepsis 203 (48.2%), asphyxia 95 (22.56%), and others 71 (16.92%) were the leading cause of neonatal admission from rst to third respectively whereas the majority 209 (53.85%) were low birth weight and the least were very low birth weight which accounts 10 (2.56 %). The majority of neonates were born from mothers whose membrane rupture from 1 to 12 hours 395 (93.58%) and those with 12 to 24 hours duration were 26 (6.15%).

Pregnancy and obstetric characteristics
The majority of maternal factors were duration of labor lasts from one to twelve hours 218 (56.41%), and the least was above twenty-four hours 19 (5.13%). Regarding labor delivery, the majority were through SVD, and least were forceps which accounts for 296 (70.26%) and 33 (7.69%) respectively.
The majority were delivered at health institution 389 (92.28 %) and the majority of presentations were cephalic which accounts for 315 (81.03%). The majority of neonates were term (37-42 weeks) 232 (54.80%) and the least were both post-term (≥42 weeks) and unknown gestational age which accounts for 13 (3.10%). The majority 385 (91.28%) had ANC follow up visits whereas 37(8.72%) did not have ANC follow up in which majority 192 (45.64%) had four ANC visits (Table 2).

Maternal health problems during pregnancy
The majority, 303 (71.9%) women did not have health problems during pregnancy while the rest 119 (28.2%) had different health problems at the latest pregnancy ( Figure 2).

Types of health problem during pregnancy
Among 119 women who had a health problem, 86 (72.27%) were with hypertensive disorders during pregnancy and those with other disorders were 2 (1.68%) (Figure 3).

Factors associated with neonatal mortality
Binary and multiple logistic regression has been used to identify associated factors. There was no marked variation by mode of delivery, presentation of labor, maternal health problems during pregnancy, and maternal educational level even though it appeared associated factors of neonatal mortality by bivariate analysis.
Neonatal gestational age group (28-32 weeks) and beyond 42 weeks were risk factors for neonatal mortality while forceps delivery was a protective factor for neonatal mortality.
Neonates with (28-32 weeks) age group were 9.5 times more likely to die than (37-42 weeks) age group. And those with gestational age group > 42 weeks were 4.6 times more likely to die than the gestational age group (37-42 weeks). Those neonates who delivered by forceps were 18% less likely to die than SVD (Table 3).

Discussion
In this study, the prevalence of neonatal mortality was 12.3%. The prevalence was higher than the national neonatal mortality with is 29 per 1000 live births (1). The study conducted at Jimma specialized hospital and Felege Hiwot Referral Hospital Ethiopia showed that neonatal mortality was 35.4% which was nearly three times higher than the prevalence in Debre Tabor Hospital and 13.29% which was comparable to this study respectively (14,17). And this study nding was higher than the study conducted in Dire Dawa Ethiopia which was 11.44% (20). The study conducted in Pakistan showed that neonatal mortality was 27% which was higher than this study (10).
This variation might be due to the different socioeconomic status of the mother and the level of differences among hospitals in which the studies conducted.
The previous studies showed that mode of delivery, educational status of secondary school, marital status, household wealth, male sex and ANC visit, preterm gestational age, and early neonatal age were associated factors for neonatal mortality (17,18). Similarly, in this study, forceps delivery and gestational age group (28-32 weeks) had a signi cant association for neonatal mortality. Besides, the gestational age group beyond 42 weeks was associated factors The neonate who was delivered through forceps were 18% less likely to die than those who delivered by spontaneous vaginal delivery. But, the study in Sudan showed that neonates delivered in C/S were more likely to die than spontaneous vaginal delivery (11). This might be due to the timely management of labor which in turn decreases labor complications.
Gestational age at delivery was signi cantly associated with neonatal mortality. The preterm newborns (28-32) weeks were 9.5 times more likely to die than those who were delivered at term (37-42) weeks which was 4 times higher than the study conducted at Bahir Dar Felege Hiwot Referral Hospital (17). This might be due to differences in hospital care levels like ANC, institutional delivery, and quality of maternal health service.

Limitation
Secondary data was used that makes it di cult to collect full data from registration and charts due to missing of some treatment outcomes and other relevant data.

Conclusion And Recommendation
The prevalence of neonatal mortality was higher compared to the national gure while Factors associated with neonatal mortality were prematurity (28-32 weeks) of gestational age, post-term (>42 weeks) of gestational age, and preventive forceps delivery. And, the way forwarded, using this stud as input would be providing client-based individualized ANC care to decrease prematurity, enhanced pre term care at NICU, and studies on causes of prematurity as it was the leading factor for neonatal mortality in this study.    Types of maternal pregnancy health problems of prevalence and associated factors of neonatal mortality among admitted neonates in NICU, Debre Tabor General Hospital, 2019.