ABSTRACT
Background and Aims
The first 28 days of life, the neonatal period, represent a phase of heightened vulnerability for infant survival globally. Despite advancements in medical interventions and increased healthcare access, a substantial number of newborns continue to succumb to preventable causes annually. This study aimed to ascertain the time to death and its associated factors among neonates admitted to the Neonatal Intensive Care Unit (NICU) at Tibebe Ghion Specialized Hospital, Ethiopia.
Methods
A retrospective follow‐up design was employed, encompassing 385 neonates admitted to the NICU of Tibebe Ghion Specialized Hospital between September 11, 2022 and September 5, 2023. Data were extracted from patient records using a standardized checklist. The outcome was time to death, with surviving neonates considered censored. Data were analyzed using STATA version 17. Kaplan–Meier survival curves analyses were used to estimate the probability of death over time. A Cox proportional hazards model was utilized to identify independent predictors of neonatal mortality. Variables with a p value < 0.05 in the multivariable Cox regression were deemed statistically significant at a 95% confidence interval.
Results
Among the 385 neonates, 12.73% experienced mortality, yielding an incidence rate of 43.80 deaths per 1000 neonatal days (95% CI: 0.03–0.06). The median time to death for neonates was 7 days (95% CI: 6.00, 8.00). After multivariable adjustment, respiratory distress syndrome (Adjusted Hazard Ratio [AHR]: 3.00; 95% CI: 1.27–7.75), congenital anomalies (AHR: 2.90; 95% CI: 1.42–5.91), very low birth weight (AHR: 2.50; 95% CI: 1.18–5.28), and gestational diabetes mellitus (AHR: 4.00; 95% CI: 1.76–9.67) were identified as significant independent predictors of neonatal death.
Conclusion
The incidence of neonatal mortality observed in this setting was notable. Respiratory distress syndrome, congenital anomalies, very low birth weight, and gestational diabetes mellitus were found to be independently associated with an increased risk of neonatal death.
Keywords: neonatal intensive care unit, neonatal mortality, survival analysis, Tibebe Ghion Specialized Hospital, time to death
1. Introduction
The first 28 days of life, the neonatal period, are the most critical for a newborn's survival globally, with most deaths occurring in the first week, often from preventable conditions like prematurity, asphyxia, and sepsis [1, 2]. In 2019, 2.4 million neonatal deaths occurred, nearly half of all under‐five mortality, primarily due to preterm birth, birth events, and infections [3]. Neonatal mortality (NM) disproportionately affects low‐ and middle‐income countries (LMICs), with 2.5 million deaths in 2018 and a projected 27.8 million by 2030 if current trends continue [4]. Sub‐Saharan Africa (SSA) has the slowest progress, accounting for 79% of neonatal deaths with South Asia [5, 6]. Alarmingly, 75% of the highest‐risk countries are in Africa, with developing nations having a nine‐fold higher NMR [4, 7]. Neonatal Intensive Care Units (NICUs) are vital for specialized care of these vulnerable newborns [8].
Neonatal mortality rates vary significantly across SSA (8–50 per 1000 live births, average 30.1) [9]. Ethiopia has made strides in child health, but neonatal mortality reduction has been modest (39 in 2000 to 33 in 2019), even increasing from 29 to 33 between 2016 and 2019 [10]. Despite global declines, 2.4 million neonatal deaths occurred in LMICs in 2020, representing 47% of under‐five deaths, with the first week being the riskiest (75%) [11]. Factors such as limited women's empowerment, low postnatal care attendance, home births, delayed breastfeeding initiation, and multiple gestations contribute to NM [12].
Addressing neonatal mortality is a global priority, as reflected in the Sustainable Development Goals (SDGs) [13]. In Ethiopia, the impact of neonatal death is profound, causing individual and familial distress, as well as hindering national progress toward health and economic development goals [14]. Studies in the Amhara region have highlighted the burden, with one study in referral hospitals reporting a 30.60% mortality rate among referred neonates, with the highest risk in the first day [15]. Analysis of EDHS 2016 data for Amhara indicated a neonatal mortality rate of 47 per 1000 live births, linked to socioeconomic factors [16]. A national meta‐analysis of 88 studies found a neonatal mortality prevalence of 16.3%, highest in Amhara (20.3%), with key predictors including gestational age, neonatal sepsis, respiratory distress syndromes (RDSs), and place of residence [17].
Current neonatal mortality rates indicate that Ethiopia is not on track to meet its National Health Sector Transformation Plan (HSTP) target (10/1000 by 2020) or the SDG target (12/1000 by 2030) [18, 19]. Despite national efforts through the HSTP [20], the first week of life remains critical, with two‐thirds of neonatal deaths potentially preventable with improved maternal and newborn care [21].
Data from Tibebe Ghion Referral Hospital's 2023 annual report revealed 206 neonatal deaths out of 2026 admissions to the NICU, highlighting a significant burden in the Amhara region [22]. While numerous studies have explored neonatal mortality globally, often focusing on prematurity, RDS, and low birth weight (LBW), research specifically examining the time to death and its determinants within the Ethiopian context remains limited, with many studies employing cross‐sectional designs.
Despite the known variations in neonatal death rates across different regions and healthcare settings, achieving national and global targets for reducing NM in Ethiopia, particularly in the Amhara region, remains a significant challenge. Notably, there is a lack of specific research conducted at Tibebe Ghion Specialized Hospital (TGSH) addressing this critical issue. This study aims to bridge this gap by investigating the time to death and its predictors among neonates admitted to the NICU at Tibebe Ghion Referral Hospital, including potentially new variables such as counseling received by mothers, to inform targeted interventions at the hospital and contribute to broader regional and national efforts to improve newborn survival.
2. Methodology and Materials
2.1. Study Design and Setting
This investigation employed a retrospective follow‐up study design to analyze the time to death and its predictors among neonates admitted to the NICU at Tibebe Ghion Referral Hospital, situated in Bahir Dar, Amhara region, Northwest Ethiopia.
Bahir Dar, the regional capital of Amhara in northern Ethiopia, had an estimated population of 751,000 in 2016. TGSH, located in Bahir Dar, functions as a key referral center for obstetric and delivery services for the city and surrounding areas. TGSH provides a comprehensive range of inpatient and outpatient services, including pharmacy, laboratory, surgical, gynecological, obstetric, maternal and child health, and emergency care, catering to over seven million individuals in the Amhara and Benishangul Gumuz regions. Affiliated with Bahir Dar University, the hospital serves as a teaching institution for medical students and residents in obstetrics and gynecology. In this medical context, fetal viability is defined at 28 weeks of gestation. Data extraction from medical records within the NICU at TGSH was conducted between May 27 and June 27, 2024.
2.2. Study Participants
The source population comprised all neonates, from birth up to 28 days of age, who were admitted to the NICU at Tibebe Ghion Referral Hospital. The study population consisted of a randomly selected cohort of neonates admitted to the NICU at Tibebe Ghion Referral Hospital between September 11, 2022 and September 5, 2023. All neonates whose medical records documented admissions between September 11, 2022 and September 5, 2023 were eligible for inclusion. Neonatal records with substantial missing data regarding key variables, specifically admission date, discharge/death date, and outcome (discharge status or death), were excluded from the analysis.
2.3. Sample Size Determination and Sampling Procedure
The target sample size for this study was calculated to be 385 neonates. This figure was determined based on the largest sample size needed to address our research objectives, primarily the estimation of neonatal mortality incidence. Utilizing a single population proportion formula (n = (Z 2 × p × (1 − p))/(d 2)), with a reported incidence (p) of 0.35 from a study conducted at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia [23], a 95% confidence interval (CI) (Z = 1.96), and a 5% margin of error (d = 0.05), the initial calculated sample size was 350. To account for potential incomplete data, a 10% contingency was added, resulting in a final sample size of 385. While sample size estimations for predictor analysis using a Cox proportional hazards model yielded smaller figures (maximum of 192), the larger sample size of 385 was selected to ensure adequate statistical power for all planned analyses.
A comprehensive list of medical records for all neonates admitted to the NICU between September 11, 2022 and September 5, 2023 (N = 2026), served as the sampling frame. The first participant was selected randomly from the first five records, and subsequently, a systematic sampling technique was employed, selecting every fifth record thereafter.
2.4. Study Variables and Operational Definitions
The outcome variable was time to neonatal death, defined as the time in days from NICU admission to death, with survivors censored. Predictor variables included:
Sociodemographic factors: Neonate's sex, neonate's age at admission, mother's age, and place of residence.
Neonatal factors: Time to first breastfeeding, perinatal asphyxia, respiratory distress, sepsis, congenital anomalies, preterm birth status, birth weight, low APGAR score, length of stay in NICU, body temperature at admission (hypothermia, hyperthermia), and hypoglycemia.
Maternal factors: Receipt of postnatal counseling, antenatal care (ANC) follow‐up, parity, gravidity, place of delivery, mode of delivery, maternal hypertension, maternal diabetes mellitus, intrapartum fever, maternal HIV status, multiple pregnancies, and gestational age.
Operational definitions included:
Neonatal death: The occurrence of death in a neonate from birth up to 28 days following admission to the NICU [24].
Follow‐up time: The duration from the time of NICU admission until either the occurrence of death or censoring.
Time to death: The interval, in days, between admission to the NICU and the date of death.
Survival time: The duration, in days, from the date of admission to the NICU until the outcome (death or censored) within the first 28 days of life.
Time scale: Measured in days from the neonate's admission to the occurrence of the outcome.
Time origin: The date of the neonate's admission to the NICU.
Complete ANC follow‐up: Attending ANC visits four or more times during the current pregnancy.
Low APGAR score: An APGAR score of < 7.
Multigravida: A recorded gravidity of two or more pregnancies.
Multiparity: A recorded parity of two or more live births.
Breastfeeding initiation time: The time elapsed until the mother first feeds breast milk to the neonate. Late initiation was defined as initiation occurring more than 1 h after birth.
Neonates' weight categories: Extremely low birth weight (ELBW) < 1000 g, very low birth weight (VLBW) 1000–1499 g, LBW 1500–2499 g, normal birth weight (NBW) 2500–3999 g, and macrosomia ≥ 4000 g.
Gestational age categories: Preterm < 37 weeks, term 37–42 weeks, and post‐term > 42 weeks, determined from the last normal menstrual period (LNMP), Ballard score, or ultrasound.
Censored: Neonates admitted to the NICU who were alive at the end of the study period or were lost to follow‐up due to discharge (home, against medical advice), or transfer to other healthcare facilities.
Event: Neonatal death.
Counseling received: Provision of information and guidance to the mother after delivery regarding newborn care, feeding practices (breastfeeding or formula feeding), early childhood development, vaccination, and hygiene [25].
2.5. Data Collection Instrument and Procedure
Initially, patient registers and charts were reviewed to assess the availability of necessary information. A structured data extraction checklist, developed in English and pretested, was used to record relevant variables from the medical records. Data collection was facilitated using the Kobo Toolbox platform for questionnaire creation and data entry. The checklist encompassed sociodemographic details, maternal history, and neonatal factors associated with neonatal mortality. Data collection was carried out by two trained nurses and one BSc‐level midwife supervisor, all with experience in the NICU.
2.6. Data Quality Assurance and Control
To ensure the quality of the collected data, continuous supervision and monitoring were implemented throughout the data collection process. A 1‐day training session was conducted for the two data collectors and the supervisor, focusing on the study objectives and appropriate data collection techniques. The supervisor reviewed the collected data daily for completeness, consistency, and accuracy, with overall oversight provided by the principal investigator. Regular support was provided to data collectors to maintain engagement and focus. Additionally, various records, including patient files and death audit reports, were cross‐referenced to verify data accuracy. The data extraction checklist was reviewed by the supervisor and principal investigator for suitability, and revisions were made following a pretest conducted on 5% of the sample at Felege Hiwot Referral Hospital. Comprehensive training on the use of the data abstraction tool and effective data collection methods was provided to all data collectors and supervisors.
2.7. Data Processing and Analysis
The collected data were coded, cleaned, edited, and entered into Kobo Toolbox before being exported to Stata version 17.0 for analysis. Descriptive statistics were computed for all independent variables and presented using text, tables, and figures. For the outcome variable, descriptive statistics such as median time with interquartile range, Kaplan–Meier survival curves, and life tables were generated. The event of interest was neonatal death, and the time to death was calculated as the number of days between admission and death.
Prior to fitting the Cox proportional hazards regression model, multicollinearity was assessed using the variance inflation factor (VIF < 10). The fundamental assumption of proportional hazards was checked using log‐log plots and the Schoenfeld residual test (global test). Both bivariable and multivariable Cox proportional hazard regression analyses were performed. Kaplan–Meier survival curves were used to estimate survival probabilities, and the log‐rank test was used to compare survival curves across different categories of independent variables. The overall model fitness was evaluated using Cox–Snell residuals.
A bivariable Cox regression model was initially used to identify candidate variables for the multivariable model (p ≤ 0.25). These candidate variables were then entered into the multivariable Cox regression model to determine the independent predictors of neonatal death. Hazard ratios (HR) with 95% CI were used to quantify the association between predictors and neonatal death. Statistical significance was defined as a p < 0.05.
3. Results
3.1. Participants Characteristics
During the period from September 11, 2022 to September 5, 2023, a total of 2026 neonates were admitted to the NICU at TGSH. Of these, records for 385 selected patients were analyzed. The duration of follow‐up for these neonates ranged from 1 to 11 days. Over half of the neonates, 217 (56.4%), were male. A significant proportion, 208 (54%), were admitted within the first week of life. Regarding birth weight at admission, 172 (44.68%) neonates weighed between 1.5 kg and 2.5 kg, with a subset of 36 (9.35%) weighing < 1500 g (Table 1).
Table 1.
Sociodemographic characteristics of neonates admitted to the NICU of TGSH from September 11, 2022 to September 5, 2023 (N = 385).
| Variables | Categories | Survival status | Total frequency | Percent (%) | |
|---|---|---|---|---|---|
| Censored (count, %) | Death (count, %) | ||||
| Age of the neonate at admission | 0–7 | 177 (85.09) | 31 (14.9) | 208 | 54.03 |
| 8–28 | 159(89.83) | 18(10.17) | 177 | 45.97 | |
| Sex of neonate | Male | 191 (88.02) | 26 (11.98) | 217 | 56.36 |
| Female | 145 (86.31) | 23 (13.69) | 168 | 43.64 | |
| Weight of the neonate at admission | 1–1.499 | 16 (44.44) | 20 (55.56) | 36 | 9.35 |
| 1.5–2.499 | 157 (91.28) | 15 (8.72) | 172 | 44.68 | |
| ≥ 2.5 | 163 (92.09) | 14 (7.91) | 177 | 45.97 | |
| Age of mother | 17–19 | 12 (70.59) | 5 (29.41) | 17 | 4.42 |
| 20–34 | 280 (90.91) | 28 (9.09) | 308 | 80.00 | |
| ≥ 35 | 44 (73.33) | 16 (26.67) | 60 | 15.58 | |
| Place of residence | Urban | 325 (87.84) | 45 (12.16) | 370 | 96.10 |
| Rural | 11 (73.33) | 4 (26.67) | 15 | 3.89 | |
The data indicated that 251 (65.19%) of the neonates' mothers were multigravida, and 244 (63.38%) were multiparous. Furthermore, 336 (87.27%) of the mothers reported receiving postnatal counseling. Among the total cohort, 338 (87.79%) mothers had four or more ANC visits during their recent pregnancy. The vast majority of neonates, 370 (96.10%), were born in a hospital setting. Regarding the mode of delivery, approximately two‐thirds of the mothers, 255 (66.23%), had spontaneous vaginal deliveries (SVD), and 117 (30.38%) underwent cesarean sections. Singleton births accounted for 355 (92.20%) of the deliveries. Obstetrical complications encountered during the recent pregnancies included pregnancy‐induced hypertension in 96 (24.93%), diabetes mellitus during pregnancy in 51 (13.25%), intrapartum fever in 62 (16.10%), and HIV infection in 4 (1.04%) mothers (Table 2).
Table 2.
Maternal characteristics of neonates admitted to the NICU of TGSH from September 11, 2022 to September 5, 2023 (N = 385).
| Variables | Categories | Survival status | Total frequency | Percent (%) | |
|---|---|---|---|---|---|
| Censored (count, %) | Death (count, %) | ||||
| Counseling received | Yes | 300 (89.29) | 36 (10.71) | 336 | 87.27 |
| No | 36 (73.47) | 13 (26.53) | 49 | 12.73 | |
| ANC follow‐up | Yes | 335 (87.47) | 48 (12.53) | 383 | 99.48 |
| No | 1 (50.00) | 1 (50.00) | 2 | 0.01 | |
| Number of ANC | No visit | 1 (50.00) | 1 (50.00) | 2 | 0.01 |
| < 4 | 29 (56.86) | 22 (43.14) | 51 | 0.13 | |
| ≥ 4 | 306 (92.17) | 26 (7.83) | 332 | 86.23 | |
| Gravidity | Primigravida | 116 (86.57) | 18 (13.43) | 134 | 34.81 |
| Multigravida | 220 (87.65) | 31 (12.35) | 251 | 65.19 | |
| Parity | Primipara | 123 (87.23) | 18 (12.77) | 141 | 36.62 |
| Multipara | 213 (87.30) | 31 (12.7) | 244 | 63.38 | |
| Place of delivery | Health institution | 325 (87.84) | 45 (12.16) | 370 | 96.1 |
| Home | 11 (73.33) | 4 (26.67) | 15 | 3.90 | |
| Types of birth | Single | 312 (87.89) | 43 (12.11) | 355 | 92.21 |
| Twin | 24 (80.00) | 6 (20.00) | 30 | 7.79 | |
| Mode of delivery | SVD | 219 (85.88) | 36 (14.12) | 255 | 66.23 |
| AID | 12 (92.31) | 1 (7.69) | 13 | 3.38 | |
| C/S | 105 (89.74) | 12 (10.26) | 117 | 30.39 | |
| PIH | Yes | 65 (67.71) | 31 (32.29) | 96 | 24.94 |
| No | 271 (93.77) | 18 (6.23) | 289 | 75.06 | |
| Intrapartum fever | Yes | 59 (95.16) | 3 (4.83) | 62 | 16.10 |
| No | 277 (85.76) | 46 (14.24) | 323 | 83.9 | |
| HIV status | Negative | 332 (87.14) | 49 (12.86) | 381 | 98.96 |
| Positive | 4 (100.00) | 0 | 4 | 1.04 | |
| Diabetes mellitus | Yes | 42 (82.35) | 9 (17.65) | 51 | 13.25 |
| No | 294 (88.02) | 40 (11.98) | 334 | 86.75 | |
Abbreviation: PIH = pregnancy‐induced hypertension.
3.2. Neonatal Characteristics and Outcomes
Among the neonates admitted to the NICU and included in the study, 365 (94.81%) had normal Apgar scores at 1 min, and 381 (98.96%) had normal scores at 5 min. Nearly half of the neonates, 183 (47.53%), presented with hypothermia upon NICU admission. Twenty‐six (6.75%) neonates required resuscitation at birth, while 198 (51.43%) did not initiate kangaroo mother care (KMC) within the first hour. Regarding breastfeeding initiation, exclusive breastfeeding was never initiated in 26 (6.75%) cases.
Based on neonatal medical conditions diagnosed at NICU admission, the prevalent complications were RDS in 159 (41.29%), neonatal sepsis in 157 (40.78%), perinatal asphyxia in 71 (18.44%), neonatal jaundice in 68 (17.66%), hypoglycemia in 76 (19.74%), and congenital anomalies in 28 (7.27%). A significant proportion, 136 (35.32%), of the neonates were born preterm (gestational age < 37 weeks). RDS and uncontrolled sepsis were identified as the most frequent causes of neonatal mortality (Table 3). The most frequently observed congenital anomalies were atrial septal defects and ventricular septal defects (Figure 1).
Table 3.
Neonatal characteristics of neonates admitted to the NICU of TGSH from September 11, 2022 to September 5, 2023 (N = 385).
| Variables | Categories | Survival status | Total frequency | Percent (%) | |
|---|---|---|---|---|---|
| Censored (count, %) | Death (count, %) | ||||
| Estimated gestational age | < 37 | 116 (85.29) | 20 (14.71) | 136 | 35.32 |
| ≥ 37 | 220 (88.35) | 29 (11.65) | 249 | 64.68 | |
| First‐minute APGAR score | < 7 | 15 (75.00) | 5 (25.00) | 20 | 5.19 |
| ≥ 7 | 321 (87.95) | 44 (12.05) | 365 | 94.81 | |
| Fifth‐minute APGAR score | < 7 | 3 (75.00) | 1 (25.00) | 4 | 1.04 |
| ≥ 7 | 333 (87.4) | 48 (12.60) | 381 | 98.96 | |
| Temperature | < 36.5 | 153 (83.61) | 30 (16.39) | 183 | 47.53 |
| 36.5–37.5 | 159 (89.83) | 18 (10.17) | 177 | 45.97 | |
| > 37.5 | 24 (96.00) | 1(4.00) | 25 | 6.49 | |
| Resuscitation at admission | Yes | 20 (76.92) | 6 (23.08) | 26 | 6.75 |
| No | 316 (88.02) | 43 (11.98) | 359 | 93.25 | |
| KMC within 1 h | Yes | 167 (89.30) | 20 (10.70) | 187 | 48.57 |
| No | 169 (85.35) | 29 (14.65) | 198 | 51.43 | |
| EBF | Yes | 317 (88.30) | 42 (11.70) | 359 | 93.25 |
| No | 19 (73.08) | 7 (26.92) | 26 | 6.75 | |
| Initiation of EBF | Not ever initiated | 19 (73.08) | 7 (26.92) | 26 | 6.75 |
| Within 1 h | 204 (92.31) | 17 (7.69) | 221 | 57.40 | |
| After 1 h | 113 (88.18) | 25 (18.12) | 138 | 35.84 | |
| Sepsis | Yes | 129 (82.17) | 28 (17.83) | 157 | 40.78 |
| No | 207 (90.79) | 21 (9.21) | 228 | 59.22 | |
| RDS | Yes | 113 (71.07) | 46 (28.93) | 159 | 41.29 |
| No | 223 (98.67) | 3 (1.33) | 226 | 58.70 | |
| Asphyxia | Yes | 57 (80.28) | 14 (19.72) | 71 | 18.44 |
| No | 279 (88.85) | 35 (11.15) | 314 | 81.56 | |
| Jaundice | Yes | 57 (83.82) | 11 (16.18) | 68 | 17.66 |
| No | 279 (88.01) | 38 (11.99) | 317 | 82.34 | |
| Hypoglycemia | Yes | 60 (78.95) | 16 (21.05) | 76 | 19.74 |
| No | 276 (89.32) | 33 (10.68) | 309 | 80.26 | |
| Congenital anomaly | Yes | 13 (46.43) | 15 (53.57) | 28 | 7.27 |
| No | 323 (90.48) | 34 (9.52) | 357 | 92.73 | |
Abbreviations: EBF = exclusive breastfeeding, RDS = respiratory distress syndrome.
Figure 1.
Congenital anomaly distribution among neonates admitted in the neonatal intensive care unit at Tibebe Ghion Referral Hospital, Bahir Dar, Amhara, Northwest Ethiopia, September 11, 2022 to September 5, 2023.
Over the study period, 282 (73.25%) of the neonates survived to the end of the follow‐up, 40 (10.39%) were lost to follow‐up, 14 (3.64%) were transferred to other healthcare facilities, and 49 (12.73%) experienced death. The overall incidence rate of mortality was 43.80 per 1000 person‐days of observation (95% CI: 0.03–0.06). The median time to death was 7.00 days (95% CI: 6.00, 8.00) (Table 4). The highest incidence of death was observed during the first 5 days of the neonatal period.
Table 4.
The Kaplan–Meier life table of time to death and its predictors among neonates admitted in neonatal intensive care unit at Tibebe Ghion Referral Hospital between September 11, 2022 and September 5, 2023 (N = 385).
| Time | At risk | Fail | Net lost | Failure function | St. error | [95% conf. int.] |
|---|---|---|---|---|---|---|
| 1 | 385 | 1 | 71 | 0.0026 | 0.0026 | 0.0004–0.0183 |
| 2 | 313 | 4 | 108 | 0.0153 | 0.0068 | 0.0064–0.0366 |
| 3 | 201 | 6 | 94 | 0.0447 | 0.0136 | 0.0246–0.0806 |
| 4 | 101 | 6 | 37 | 0.1015 | 0.0258 | 0.0612–0.1657 |
| 5 | 58 | 12 | 18 | 0.2874 | 0.0520 | 0.1992–0.4035 |
| 6 | 28 | 7 | 5 | 0.4655 | 0.0702 | 0.3400–0.6112 |
| 7 | 16 | 4 | 2 | 0.5992 | 0.0782 | 0.4521–0.7507 |
| 8 | 10 | 4 | 1 | 0.7595 | 0.0778 | 0.5987–0.8918 |
| 9 | 5 | 3 | 0 | 0.9038 | 0.0612 | 0.7471–0.9815 |
| 10 | 2 | 1 | 0 | 0.9519 | 0.0458 | 0.8063–0.9963 |
| 11 | 1 | 1 | 0 | 1.0000 | — | — |
Note: Net lost equals the number lost minus the number who entered.
The incidence rate of death was higher among female neonates (48.70 per 1000 neonatal days, 95% CI: 0.032–0.073) compared to male neonates (40.10 per 1000 person‐days, 95% CI: 0.03–0.06). Similarly, neonates admitted at < 8 days of age had a higher incidence rate (50.60 per 1000 neonatal days, 95% CI: 0.04–0.07) compared to those admitted at 8 days or older (35.40 per 1000 person‐days, 95% CI: 0.02–0.06). The incidence rate was also higher among neonates from rural areas (80 per 1000 neonatal days, 95% CI: 0.04–0.07) compared to urban neonates (42.10 per 1000 person‐days, 95% CI: 0.02–0.06).
The median time to death was shorter for female neonates (6.00 days, 95% CI: 6.00, 7.00) compared to males (8.00 days, 95% CI: 6.00, 9.00). Neonates admitted at < 8 days had a shorter median time to death (6.00 days, 95% CI: 6.00, 8.00) compared to those admitted at 8 days or older (7.00 days, 95% CI: 5.00, 8.00). Similarly, neonates weighing < 1.5 kg had a shorter median time to death (6.00 days, 95% CI: 5.00, 8.00) compared to those weighing between 1.5 kg and 2.499 kg (7.00 days, 95% CI: 6.00, 8.00).
The Kaplan–Meier failure estimate curve indicated a rapid increase in the probability of neonatal death during the first 7 days of follow‐up, suggesting a heightened risk of mortality in this period. The curve also showed an initially elevated hazard of death between Days 5 and 7 (Figure 2).
Figure 2.
Overall Kaplan–Meier failure estimate with 95% confidence intervals of neonates admitted to NICU in Tibebe Ghion Specialized Hospital between September 11, 2022 and September 5, 2023.
3.3. Predictors of Mortality Among Neonates Admitted to NICU
Bivariable and multivariable Cox proportional hazard models were utilized to identify factors influencing the time to death and mortality of neonates admitted to the NICU. The bivariable Cox regression analysis revealed several independent variables significantly associated with neonatal mortality and time to death (Table 5). These included RDS, congenital anomalies, neonatal birth weight, pregnancy‐induced hypertension, and gestational diabetes mellitus (GDM), all exhibiting p values < 0.250 and thus qualifying for inclusion in the multivariable analysis.
Table 5.
Bivariable and multivariable Cox‐regression of time to death and its predictors among neonates admitted in the neonatal intensive care unit at Tibebe Ghion Specialized Hospital between September 11, 2022 and September 5, 2023.
| Variable | Categories | Survival status | CHR (95% CI) | p | AHR (95% CI) | |
|---|---|---|---|---|---|---|
| Censored (count, %) | Death (count, %) | |||||
| PIH | Yes | 65 (67.71) | 31 (32.29) | 2.53 (1.35–4.72)* | 0.004 | 1.70 (0.85–3.36) |
| No | 271 (93.77) | 18 (6.23) | 1 | |||
| GDM | Yes | 42 (82.35) | 9 (17.65) | 2.20 (1.04–4.64)* | 0.038 | 4.00 (1.76–9.67)** |
| No | 294 (88.02) | 40 (11.98) | 1 | |||
| RDS | Yes | 113 (71.07) | 46 (28.93) | 4.20 (1.74–9.93)* | 0.001 | 3.00 (1.27–7.75)** |
| No | 223 (98.67) | 3 (1.33) | 1 | |||
| Congenital anomaly | Yes | 13 (46.43) | 15 (53.57) | 2.53 (1.34–4.78)* | 0.004 | 2.90 (1.42–5.91)** |
| No | 323 (90.48) | 34 (9.52) | 1 | |||
| Neonatal weight | ≥ 2.5 | 163 (92.09) | 14 (7.91) | 1 | ||
| 1.5–2.499 | 157 (91.28) | 15 (8.72) | 1.40 (0.65–2.84)* | 0.409 | 1.60 (0.77–3.49)** | |
| 1–1.499 | 16 (44.44) | 20 (55.56) | 2.00 (1.002–4.173) | 0.049 | 2.50 (1.18–5.28) | |
Note: 1—reference category.
Abbreviations: GDM = gestational diabetes mellitus, PIH = pregnancy‐induced hypertension, RDS = respiratory distress syndrome.
Statistically significant at bivariable with 0.25 level significant.
Statistically significant at multivariable with 0.05 level significant.
The final multivariable Cox proportional hazard regression analysis identified RDS, congenital anomalies, neonatal birth weight, and GDM as independent predictors of neonatal death at a 5% significance level. The Schoenfeld residuals test confirmed that the Cox proportional hazard assumption was met for each covariate and for the overall model (global test p = 0.229). Graphical assessment of the covariates also supported the proportional hazards assumption. Based on the Cox–Snell residuals, the Nelson–Aalen hazard plot showed a close alignment with the 45° reference line, indicating a good fit of the model (Figure 3).
Figure 3.
Cox–Snell residual Nelson–Alen cumulative hazard graph on neonates admitted in NICU on neonates admitted in NICU at TGSH from September 11, 2022 to September 5, 2023.
The multivariable Cox proportional hazard regression analysis showed that neonates with RDS had a threefold higher hazard of mortality compared to those without RDS (Adjusted Hazard Ratio [AHR] = 3.00, 95% CI: 1.27–7.75). The hazard of death for neonates with congenital anomalies was 2.90 times higher compared to those without such anomalies (AHR = 2.90, 95% CI: 1.42–5.91). Furthermore, neonates with LBW had a 2.5 times higher hazard of death compared to those with NBW (AHR = 2.50, 95% CI: 1.18–5.28). Similarly, neonates born to mothers with GDM had a fourfold increased hazard of death (AHR = 4.00, 95% CI: 1.76–9.67) compared to those born to mothers without GDM (Table 5).
4. Discussion
This study aimed to determine the time to death and its predictors among neonates admitted to the NICU of TGSH. A total of 385 neonates admitted to the NICU were included in this retrospective analysis. The study revealed that 49 neonates died during the 1120 person‐days of follow‐up, with a median time to death of 7 days (IQR: 5.00–8.00 days). The incidence density rate of mortality among admitted neonates was 43.80 per 1000 neonate‐days (95% CI: 0.03–0.058). This incidence rate is higher than that reported in a meta‐analysis of studies in SSA, which found an incidence rate of 24.53 per 1000 person‐days among neonates in NICUs [26]. This discrepancy might be attributed to the pooled nature of the SSA study, which synthesized findings from diverse settings and methodologies, potentially masking the higher risk in specific hospital environments like TGSH.
Our findings also indicate a higher incidence rate compared to previous studies within Ethiopia, such as those conducted at Arba Minch General Hospital (31.60 per 1000 neonate‐days) [27], Gondar Comprehensive Specialized Hospital (34.69 per 1000 neonatal admission days) [23], Amhara regional State referral hospitals (27.00 per 1000 neonatal days) [28], and the Jabitehnan district (1.64 per 1000 days) [29]. Furthermore, the incidence rate observed in Wolayita Sodo (77.00 neonatal deaths per 1000 neonate‐days) [30] was even higher than in our study. These variations could be due to differences in sample size, study design, study period, and the level of healthcare service coverage across these regions and hospitals.
The median time to death in our study indicates that approximately 60.% of neonatal deaths occurred within the first 7 days, and about 76% within the first 8 days, with roughly 29% occurring within the initial 5 days of follow‐up. This aligns with findings from a multicenter study in Nigeria and Kenya [31], as well as studies at Wollega University Referral Hospital and Nekemte Specialized Hospital [32], and the University of Gondar Comprehensive Specialized Hospital [23]. The elevated risk of death in the early neonatal period is often linked to complications arising from preterm birth, including RDS, infections, and congenital anomalies, as well as LBW, inadequate prenatal care, maternal health issues (such as hypertension or diabetes), and complications during childbirth. Timely and appropriate medical interventions, coupled with improved access to quality prenatal and neonatal care, are crucial for mitigating neonatal mortality during this critical phase.
Interestingly, and in contrast to existing literature, this study found a higher incidence rate and shorter median time to death among female neonates, despite the fact that more than half of the admitted neonates were male. Although biologically male neonates are often more vulnerable to neonatal complications such as RDS, our findings may reflect context‐specific factors. These could include differences in illness severity at admission, potential delays in care‐seeking behavior for female neonates, or gender‐based disparities in postnatal care and attention. Additionally, the smaller number of female neonates in our cohort may have amplified the proportional mortality rate in this group. These factors may help explain the observed reversal in expected trends, though further prospective research is needed to explore gender‐specific care patterns and outcomes in similar settings.
RDS significantly increased the hazard of neonatal death in this study, with affected neonates having a threefold higher risk compared to those without RDS. This finding is consistent with research from tertiary hospitals in Addis Ababa, Jimma Medical Center, Felege‐Hiwot Hospital, and the Hadiya zone [24, 33, 34, 35]. The underlying reason is the immaturity of the lungs in newborns with RDS, leading to breathing difficulties and impaired oxygen exchange, potentially resulting in respiratory failure and death. Specialized hospitals may also manage a higher volume of premature and high‐risk births, further elevating the incidence of RDS and its associated mortality risks. Early detection, prompt medical management, and appropriate respiratory support are essential for improving outcomes in these cases.
Our study also found that neonates with congenital anomalies had a 2.9 times higher risk of mortality compared to those without these conditions. This is in agreement with studies from Felege‐Hiwot Hospital, Hawassa University Comprehensive Specialized Hospital, and Ayder Comprehensive Specialized Hospital [34, 36, 37]. Congenital anomalies, ranging from structural defects to genetic disorders, can significantly impact a newborn's health and survival. The need for specialized care and interventions for these neonates, which may not be universally accessible, could contribute to the increased mortality risk observed in specialized hospital settings where more complex cases are concentrated.
Neonates with a birth weight below 1500 g in our study had a 2.50 times higher hazard of death compared to those with NBW. This is supported by evidence from the Amhara Regional State, Felege‐Hiwot Hospital, Bombe Primary Hospital in southern Ethiopia, Hawassa University Comprehensive Specialized Hospital, and Ayder Comprehensive Specialized Hospital [15, 34, 38, 39]. VLBW neonates are more susceptible to complications such as RDS and infections due to their underdeveloped organ systems and immune function. The intensive medical care and monitoring required by these infants can strain the resources of specialized hospitals. Furthermore, LBW is associated with a higher risk of long‐term developmental issues, increasing their vulnerability.
GDM in the mother was another significant predictor of neonatal mortality, with neonates born to mothers with GDM having a fourfold higher hazard of death. This finding aligns with a study conducted at Black Lion Specialized Hospital in Addis Ababa [40]. Gestational diabetes can lead to various complications in newborns, including macrosomia, hypoglycemia, RDS, and jaundice, all of which can increase the risk of neonatal death, particularly in specialized hospitals managing high‐risk pregnancies. Effective management of diabetes during pregnancy is crucial for reducing these adverse outcomes for both mother and child.
5. Limitations of the Study
The reliance on secondary data sources may have resulted in the absence of potentially relevant predictors not consistently documented in patient charts. Consequently, factors such as maternal nutritional status, educational attainment, family‐related dynamics, and socioeconomic status could not be comprehensively addressed within this study. Furthermore, the exclusion of charts with incomplete data may have introduced a degree of selection bias.
6. Conclusion
The median time to death was determined to be 7 days. The overall incidence rate of neonatal mortality was 43.80 per 1000 neonatal days, with a higher proportion of deaths occurring within the first 5 days of life. Several factors, including RDS, congenital anomalies, neonatal birth weight, and GDM, were identified as significant predictors of neonatal mortality. Close monitoring and targeted interventions addressing these factors are crucial for optimizing care and outcomes for both mothers and their newborns, particularly concerning congenital anomalies, birth weight, gestational diabetes, and the management of RDS.
7. Recommendations
To reduce neonatal mortality at TGSH, comprehensive efforts are crucial. This includes prioritizing and enhancing focused ANC to identify and intervene in potential risks early, with specific attention to high‐risk pregnancies. Ensuring appropriate medical interventions and treatments for conditions like RDS and GDM is vital, as is providing adequate and specialized neonatal care for newborns with congenital anomalies.
To effectively address the significant burden of neonatal mortality in the Amhara region, the Amhara Regional Health Bureau should adopt a multifaceted approach targeting preventable predictors. Initiatives should focus on actively engaging and empowering mothers to participate in and receive quality care, particularly those with GDM and for neonates with congenital anomalies, RDS, and LBW.
For future research, given that this study relied on documented data which may have limitations, prospective longitudinal studies are recommended. These studies should aim to incorporate vital variables not consistently available in secondary data, such as maternal education status, socioeconomic factors, and other potentially influential elements to provide a more comprehensive understanding of neonatal mortality.
Author Contributions
Gebiyaw Wudie Tsegaye: conceptualization, investigation, writing – original draft, methodology, validation, visualization, writing – review and editing, software, formal analysis, data curation, supervision. Maru Birhanu Addis: conceptualization, investigation, methodology, visualization, writing – original draft, formal analysis, data curation. Kindie Fentahun Muchie: investigation, writing – review and editing, methodology, validation, software, supervision, data curation. All authors have read and approved the final version of the manuscript.
Funding
The authors have nothing to report.
Disclosure
The lead author Gebiyaw Wudie Tsegaye affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Ethics Statement
Ethical approval was obtained from the IRB of CMHS, Bahir Dar University with a Protocol Number of 940/2024. A formal letter was obtained from Bahir Dar University to proceed with the field work and from TGRH to access the medical records.
Consent
Confidentiality was guaranteed by excluding names or any other personal identifiers from the data collection checklists.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgments
The authors would like to express their gratitude to the College of Medicine and Health Science, School of Public Health, Bahir Dar University, Bahir Dar, Ethiopia. They are also grateful to data collectors, supervisors, and Tibebe Ghion Specialized Hospital NICU Ward administration and card‐room staff for their valuable contribution to the accomplishment of this study.
Data Availability Statement
The data were collected from patient charts of individual neonates admitted in the NICU at TGSH. The collected information was used only for research purposes and was kept confidential. The corresponding author will provide the data set used and or analyzed during the study upon responsible request. The corresponding author (Gebiyaw Wudie Tsegaye) had full access to all of the data in this study and took complete responsibility for the integrity of the data and the accuracy of the data analysis.
References
- 1. World Health Organization , Reaching the Every Newborn National 2020 Milestones: Country Progress, Plans and Moving Forward (WHO, 2017).
- 2. Roro E. M., Tumtu M. I., and Gebre D. S., “Predictors, Causes, and Trends of Neonatal Mortality at Nekemte Referral Hospital, East Wollega Zone, Western Ethiopia (2010–2014). Retrospective Cohort Study,” PLoS One 14, no. 10 (2019): e0221513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. WHO , Newborns: Improving Survival and Well‐Being (WHO, 2021), https://www.who.int/news-room/fact-sheets/detail/newborns-reducing-mortality.
- 4. Hug L., Alexander M., You D., and Alkema L., “National, Regional, and Global Levels and Trends in Neonatal Mortality Between 1990 and 2017, With Scenario‐Based Projections to 2030: A Systematic Analysis,” Lancet Global Health 7, no. 6 (2019): e710–e720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Akombi B. J. and Renzaho A. M., “Perinatal Mortality in Sub‐Saharan Africa: A Meta‐Analysis of Demographic and Health Surveys,” Annals of Global Health 85, no. 1 (2019): 106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Dhaded S. M., Somannavar M. S., Vernekar S. S., et al., “Neonatal Mortality and Coverage of Essential Newborn Interventions 2010‐2013: A Prospective, Population‐Based Study From Low‐Middle Income Countries,” Reproductive Health 12 (2015): S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mihiretu A., Negash T., and Elazar T., “Perinatal Death and Associated Factors in Wolaita Sodo Referral Hospital, Southern Ethiopia: A Facility Based Cross‐Sectional Study,” Journal of Primary Health Care 7, no. 2 (2017): 1–5. [Google Scholar]
- 8. Chow S., Chow R., Popovic M., et al., “A Selected Review of the Mortality Rates of Neonatal Intensive Care Units,” Frontiers in Public Health 3 (2015): 225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ethiopian Public Health Institute (EPHI) , ICF. Ethiopia Mini Demographic and Health Survey 2019: key Indicators. Addis Abeba, Ethiopia and Rockville Mary land, USA: EPHI and ICF; 2019.
- 10. Girma D., Dejene H., and Adugna L., “Predictors of Neonatal Mortality in Ethiopia: A Comprehensive Review of Follow‐Up Studies,” International Journal of Pediatrics 2022 (2022): 1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Al‐Selwi A. and Barkat A., “International and National Data on Maternal, Neonatal, and Child Mortality Rates and Policies to Reduce the High in the Middle East and Morocco: A Multilevel Meta‐Analysis,” Journal of Bioscience and Applied Research 9, no. 4 (2023): 261–291. [Google Scholar]
- 12. Mitiku H. D., “Neonatal Mortality and Associated Factors in Ethiopia: A Cross‐Sectional Population‐Based Study,” BMC Women's Health 21, no. 1 (2021): 156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tefera Y. G. and Ayele A. A., “Newborns and Under‐5 Mortality in Ethiopia: The Necessity to Revitalize Partnership in Post‐COVID‐19 Era to Meet the SDG Targets,” Journal of Primary Care & Community Health 12 (2021): 2150132721996889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. UNICEF Ethiopia , Neonatal Mortality (UNICEF Ethiopia, 2020), https://www.unicef.org/ethiopia/child-survival-neonatal-mortality.
- 15. Yeshaneh A., Tadele B., Dessalew B., et al., “Incidence and Predictors of Mortality Among Neonates Referred to Comprehensive and Specialized Hospitals in Amhara Regional State, North Ethiopia: A Prospective Follow‐Up Study,” Italian Journal of Pediatrics 47, no. 1 (2021): 186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Wolde H., “Determinants of Neonatal Mortality in Amhara Regional State, Ethiopia: Further Investigation of EDHS 2016,” Journal of Clinical Images and Medical Case Reports 4, no. 9 (2023): 2594. [Google Scholar]
- 17. Aynalem Y. A., Shiferaw W. S., Akalu T. Y., Dargie A., Assefa H. K., and Habtewold T. D., “The Magnitude of Neonatal Mortality and Its Predictors in Ethiopia: A Systematic Review and Meta‐Analysis,” International Journal of Pediatrics 2021 (2021): 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ethiopian Federal Ministry of Health , Health Sector Transformation Plan (Ethiopian Federal Ministry of Health, 2015), 184.
- 19. United Nations , Transforming Our World: The 2030 Agenda for Sustainable Development (United Nations, 2015), http://www.un.org.
- 20. Weldemarium T., Kebede Z., Ababor S., Bogale F., Solomon D., and Gebreyohannes Y., “Reducing Neonatal Mortality in Ethiopia: A Call for Urgent Action! An Evidence Brief for Policy,” Population Medicine 5 (2023): A665. [Google Scholar]
- 21. Bale J. R., Stoll B. J., and Lucas A. O., Reducing Neonatal Mortality and Morbidity. Improving Birth Outcomes: Meeting the Challenge in the Developing World (National Academies Press, 2003). [PubMed] [Google Scholar]
- 22. Tibebe Ghion Referral Hospital , Annual Report of Neonatal Death (TGRH, 2023).
- 23. Gudayu T. W., Zeleke E. G., and Lakew A. M., “Time to Death and Its Predictors Among Neonates Admitted in the Intensive Care Unit of the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia,” Research and Reports in Neonatology 10 (2020): 1. [Google Scholar]
- 24. Menalu M. M., Gebremichael B., Desta K. W., et al., “Time to Death and Its Predictors Among Neonates Who Were Admitted to the Neonatal Intensive Care Unit at Tertiary Hospital, Addis Ababa, Ethiopia: Retrospective Follow Up Study,” Frontiers in Pediatrics 10 (2022): 913583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. WHO , Kangaroo Mother Care: A Practical Guide (WHO, 2021), https://www.who.int/maternal_child_adolescent/documents/care_of_small_babies/en/.
- 26. Bitew Z. W., Alemu A., Ayele E. G., Jember D. A., Haile M. T., and Worku T., “Incidence Density Rate of Neonatal Mortality and Predictors in Sub‐Saharan Africa: A Systematic Review and Meta‐Analysis,” International Journal of Pediatrics 2020 (2020): 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Dessu S., K M., Gebremeskel F., et al., “Predictors of Neonatal Mortality Among Neonates Who Admitted in Neonatal Intensive Care Unit at Arba Minch General Hospital,” Ethiopian Journal of Health Development 33, no. 1 (2019): 40–47. [Google Scholar]
- 28. Mengistu B. A., Yismaw A. E., Azene Z. N., et al., “Incidence and Predictors of Neonatal Mortality Among Neonates Admitted in Amhara Regional State Referral Hospitals, Ethiopia: Prospective Follow Up Study,” BMC Pediatrics 20 (2021): 142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Dessie A. M., Nigatu D., Yadita Z. S., et al., “Neonatal Mortality and Its Predictors Among Neonates in Jabitehnan District, Northwest Ethiopia: A Single‐Arm Retrospective Cohort Study,” Health Science Reports 6, no. 10 (2023): e1613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Orsido T. T., Asseffa N. A., and Berheto T. M., “Predictors of Neonatal Mortality in Neonatal Intensive Care Unit at Referral Hospital in Southern Ethiopia: A Retrospective Cohort Study,” BMC Pregnancy and Childbirth 19 (2019): 83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Nabwera H. M., Wang D., Tongo O. O., et al., “Burden of Disease and Risk Factors for Mortality Amongst Hospitalized Newborns in Nigeria and Kenya,” PLoS One 16, no. 1 (2021): e0244109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Tolossa T., Wakuma B., Mengist B., et al., “Survival Status and Predictors of Neonatal Mortality Among Neonates Admitted to Neonatal Intensive Care Unit (NICU) of Wollega University Referral Hospital (WURH) and Nekemte Specialized Hospital, Western Ethiopia: A Prospective Cohort Study,” PLoS One 17, no. 7 (2022): e0268744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Seid S. S., Ibro S. A., Ahmed A. A., et al., “Causes and Factors Associated With Neonatal Mortality in Neonatal Intensive Care Unit (NICU) of Jimma University Medical Center, Jimma, South West Ethiopia,” Pediatric Health, Medicine and Therapeutics 10 (2019): 39–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Woelile T. A., Kibret G. T., Workie H. M., et al., “Survival Status and Predictors of Mortality Among Low‐Birth‐Weight Neonates Admitted to the Neonatal Intensive Care Unit at Felege Hiwot Comprehensive Specialized Hospital, Bahir Dar, Ethiopia, 2020,” Pediatric Health, Medicine and Therapeutics 12 (2021): 451–466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Moges S., Mekango D. E., and Astatkie A., “Mortality and Its Predictors Among Neonates Admitted to a Neonatal Intensive Care Unit in Hadiya Zone, Southern Ethiopia: A Retrospective Cohort Study,” Journal of Pediatrics and Neonatology 2 (2021): 10–16. [Google Scholar]
- 36. Taye K., Kebede Y., Tsegaw D., and Ketema W., “Predictors of Neonatal Mortality Among Neonates Admitted to the Neonatal Intensive Care Unit at Hawassa University Comprehensive Specialized Hospital, Sidama Regional State, Ethiopia,” BMC Pediatrics 24, no. 1 (2024): 237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Hadgu F. B., Gebretsadik L. G., Mihretu H. G., et al., “Prevalence and Factors Associated With Neonatal Mortality at Ayder Comprehensive Specialized Hospital, Northern Ethiopia. A Cross‐Sectional Study,” Pediatric Health, Medicine and Therapeutics 11 (2020): 29–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Berhanu B., Oljira L., Demana M., et al., “Survival and Predictors of Mortality Among Neonates Admitted to Neonatal Intensive Care Unit at Bombe Primary Hospital, Southern Ethiopia: Institution‐Based Retrospective Cohort Study,” Pediatric Health, Medicine and Therapeutics 12 (2021): 239–249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Kahsay A. H., Abebe H. T., Gebretsadik L. G., et al., “Survival and Predictors of Early Neonatal Death in Neonatal Intensive Care Unit of Mekelle General and Ayder Comprehensive Specialized Hospitals, Northern Ethiopia, 2018: Prospective Cohort Study,” preprint, Research Square, 2019.
- 40. Aynalem Y. A., Mekonen H., Getaneh K., et al., “Determinants of Neonatal Mortality Among Preterm Births in Black Lion Specialized Hospital, Addis Ababa, Ethiopia: A Case–Cohort Study,” BMJ Open 12, no. 2 (2022): e043509. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data were collected from patient charts of individual neonates admitted in the NICU at TGSH. The collected information was used only for research purposes and was kept confidential. The corresponding author will provide the data set used and or analyzed during the study upon responsible request. The corresponding author (Gebiyaw Wudie Tsegaye) had full access to all of the data in this study and took complete responsibility for the integrity of the data and the accuracy of the data analysis.