You are here

  • Home
  • Archive
  • Volume 14, Issue 4
  • Survival status and predictors of mortality among preterm neonates admitted in Bench Sheko Zone, Sheka Zone and Keffa Zone Governmental Hospitals, Southwest Ethiopia (2021): prospective follow-up study

Article Text

Article menu
Paediatrics
Original research
Survival status and predictors of mortality among preterm neonates admitted in Bench Sheko Zone, Sheka Zone and Keffa Zone Governmental Hospitals, Southwest Ethiopia (2021): prospective follow-up study
  1. Esmelealem Mihretu1,
  2. Yalemtsehay Dagnaw Genie2,
  3. Emebet Adugnaw3,
  4. Aster Tadesse Shibabaw4
  1. 1 Pediatrics and Child Health Nursing, Debre Markos University, Debre Markos, Ethiopia
  2. 2 Nursing, Mizan-Tepi University, Mizan Teferi, Ethiopia
  3. 3 Department of Public Health, Mizan-Tepi University, Mizan Teferi, Ethiopia
  4. 4 Debre Markos University, Debre Markos, Amhara, Ethiopia
  1. Correspondence to Dr Esmelealem Mihretu; esmemahi2119{at}gmail.com

Abstract

Introduction Prematurity presents a significant challenge to the global community due to the rapid increase in its incidence and its disproportionate contribution to increased infant mortality rates.

Objective To assess the survival status and predictors of mortality among preterm neonates.

Design A multicentre prospective follow-up study was used.

Setting 625 preterm neonates were admitted to hospitals for secondary level of care. The study covers the Bench Maji Zone, Keffa Zone, Sheka Zone, nearby woredas and portions of the Gambella area in Southwest Ethiopia.

Participants 614 preterm neonates with gestational age less than 37 weeks were entered for follow-up and 400 neonates were censored. Neonates with severe fetal malformations and neonates who need urgent referral were excluded from the study.

Results Overall, 200 (32.57%) participants died with an incidence rate of 61.69 deaths per 1000 person-day observations (95% CI: 53.71 to 70.86). Poor kangaroo mother care (KMC) services (adjusted HR (AHR)=0.19, 95% CI: 0.12 to 0.29), sex (AHR=0.66, 95%, CI: 0.47 to 0.94), not initiating breast feeding (HR=2.78, 95% CI: 1.8 to 4.28), hypothermia (AHR=0.63, 95% CI: 0.44 to 0.92), anaemia (AHR=6.2, 95% CI: 2.34 to 16.43) and gestational age less than 28 weeks (AHR=9.28, 95% CI: 1.78 to 48.42) were independent predictors.

Conclusion and recommendation The rate of preterm neonatal mortality was high compared with the Ethiopia Demographic and Health Survey report nationally. Healthcare workers should encourage KMC services and breastfeeding initiation and prevent preterm neonates from being anaemic to increase their chances of survival.

  • Neonatal intensive & critical care
  • NEONATOLOGY
  • Nurses
  • Paediatric intensive & critical care
  • Nursing Care

Data availability statement

Data are available upon reasonable request. all data are incorporated within the manuscript

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi-org.ezproxy.u-pec.fr/10.1136/bmjopen-2024-083897

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    STRENGTHS AND LIMITATIONS OF THIS STUDY

    • The study did address the probable service-related predictors of mortality.

    • The study was conducted in a multicentre setting using random sampling and patients were followed for a long duration with a maximum of 28 days, increasing the study’s generalisation.

    • Data on the site of delivery, home versus the hospital, might be incomplete.

    • Possibility of medical record errors and diagnostic subjectivity was not controlled.

    Introduction

    Preterm is defined as babies being born alive before 37 weeks of pregnancy are completed. It has subcategories based on gestational age (GA): extremely preterm (less than 28 weeks), very preterm (28–32 weeks) and moderate-to-late preterm (32–37 weeks).1 2

    In addition, preterm birth is a pervasive disorder that impacts all the functioning of the neonates who survived both in the short and long term, such as neurodevelopment, education, psychosocial, growth and health outcomes. Mortality is also related to GA, and preterm birth continues to be the leading cause of perinatal and postnatal mortality mainly in undeveloped countries where health services are limited and not functioning well.3

    Prematurity presents a significant challenge to the global community because of the rapid increase in its incidence and disproportionate contribution to increased infant mortality rates.1 Globally, each year, 15 million babies are born preterm (before 37 completed weeks of gestation), which is estimated to be approximately 11% of all deliveries.4 Although the burden of under-5 mortality decreased between 2010 (7.6 million)5 and 2015 (5.942 million),6 the percentage of mortality caused by preterm-related complications has increased from 14.1%5 to 17.8%,6 and more than 80% of perinatal problems and mortality from preterm deliveries are reported. Among all neonatal deaths globally in 2013, 35% were caused by preterm birth complications alone.7 The rate of preterm neonatal mortality increased in sub-Saharan Africa from 0.326 million in 20138 to 0.356 million in 2015.6 Survivors experience inferior neurological development, poor academic performance, higher risk of cerebral palsy and metabolic problems in adulthood after surviving the neonatal period.9

    Preterm labour is the main cause of neonatal deaths in many regions in Ethiopia, and neonates delivered with shorter GAs have demonstrated higher neonatal mortality than term babies.10 More than two-thirds (76%) of the neonatal deaths in the southwest area of Ethiopia, which accounted for 22.8% of all neonatal deaths, were due to preterm birth.11 There is a need to identify the causes of preterm mortality given the growing impact of neonatal deaths on overall child mortality.6 12 To a certain extent, researchers have been more interested in trends concerning the occurrence and subsequent mortality of preterm births over time than in the rates and factors affecting the survival of preterm infants. There is still more research to be conducted on the direct evidence indicating an increased tendency for preterm birth-related mortality. For prenatal care evaluation, parental education, careful selection and development of clinical standards, current estimates of the survival status of preterm neonates admitted to the neonatal intensive care unit (NICU) are required. Thus, the goal of this study was to assess survival status and identify the predictors of mortality among preterm neonates at selected governmental hospitals in Southwest Ethiopia.

    Methods

    Study design and setting

    Preterm neonates admitted to the NICU wards at Mizan-Tepi University Teaching Hospital, Tepi General Hospital and G/Tsadik Shawo General Hospital located in Bench Sheko, Sheka and Keffa Zones, Southwest Ethiopia were the subjects of this institution-based prospective follow-up study from 10 March to 30 September 2020. They cover the Bench Maji Zone, Keffa Zone, Sheka Zones, nearby woredas and portions of the Gambella area, with a catchment population of more than 2.7 million (Majang Zone). In these three hospitals, more than 665 infants are born each month, with 70 of those admitted to the NICU due to being preterm.

    Study population

    All preterm neonates who are admitted to the intensive care unit of Bench Sheko Zone, Keffa Zone and Sheka Zone hospitals, Southwest Ethiopia during the study period.

    Inclusion criteria

    Mothers with their preterm neonates whose GA was less than 37 weeks, who were admitted in the selected hospitals during the study period, were included.

    Exclusion criteria

    Neonates with severe fetal malformations, mothers who were unable to speak, neonates who need urgent referral and mothers with psychiatric illnesses were excluded from the study.

    Sample size and sampling procedure

    Sample size was determined by using STATA statistical package (Cox model), V.14 based on the following assumptions: HR of 1.55 for perinatal asphyxia that gives a maximum sample size among covariates,13 a variability (SD) of 0.5, probability of failure (event) of 0.288, a 5% margin of error, 95% CI and 80% power. The required number of events was 164, and the number of outcomes was 568. The final sample size was 625 after a 10% non-response rate was added.

    The total number of preterm neonates admitted to the NICU on a monthly basis in each of the chosen hospitals was assessed before the data collection began. The overall sample size of the study was then proportionally distributed to each chosen hospital depending on the number of preterm neonates admitted to the NICU in the past. Then, at each hospital’s corresponding NICU, we included all preterm neonates who had been admitted. Study participants were selected from each hospital using a sequential sampling method. All recruited study subjects were tracked until the desired result (death or censure) manifested itself.

    Measurement and variables

    Survival status of the preterm neonates was the outcome variable in this study. Sociodemographic-related factors (date of birth, date of admission, sex of the neonate, maternal age, age of the neonate at admission, residency of the mother and place of delivery), neonatal-related factors (birth weight, Apgar score, weight for gestational age at birth, feeding status, breathing condition at birth and kangaroo mother care (KMC)), neonatal preterm-related medical and surgical complications factors (sepsis, necrotising enterocolitis, intraventricular haemorrhage, asphyxia, respiratory distress syndrome (RDS), jaundice, pulmonary haemorrhage, pulmonary hypertension, hypoxic ischaemic encephalopathy, anaemia, congenital anomalies, hypothermia and hypoglycaemia), maternal and obstetric-related factors (maternal chronic disease, obstetric complications, pregnancy-induced hypertension, maternal exposure status to corticosteroids before delivery, presentation at delivery, mode of delivery, antenatal care status, parity of the mother and pregnancy type), and institutional and professional-related factors (place of KMC service provided, level of NICU, hospital level and availability of resuscitation equipment) were the independent variables.

    Operational definitions

    Survival time: the time from admission to the NICU until the occurrence of an event (death) during the study period.

    Event: death of a preterm neonate after admission to the NICU during the hospitalisation period.

    Censored: preterm neonates who had survived during the follow-up period (including preterm neonates who were discharged with improvement, lost to follow-up (left against medical advice or transferred to other health institutions), withdrawn (if the mother refused the follow-up due to inconvenience) and still stayed with admission beyond 28 days of neonatal age).

    Time scale: the survival time measured in days.

    Starting time: the first day of admission of the preterm neonate at the NICU.

    End of follow-up time: the last day of an event or censored occurrence in the hospital.

    Preterm neonates: neonates who were born less than 37 weeks of GA, irrespective of birth weight.13

    Data collection tools and procedure

    Data were gathered using a pretested interviewer-administered questionnaire based on pertinent literature.10 13–22 The questionnaire was prepared in English and translated into Amharic and then retranslated into English by senior experts. Sociodemographic, neonatal, medical and surgical complications related to preterm birth, maternal and obstetric-related, and health institution-related variables were included.

    Data were gathered using in-person interview for primary data and chart review for secondary data. Secondary data were used to obtain baseline data of mothers that were unable to be collected. After putting aside all baseline information on sociodemographic factors, obstetrics-related factors and neonatal factors related to the date of birth and the first day of enrolment, daily follow-up assessments of factors that could occur beginning on the first day of follow-up (ie, from the time of admission to the neonatal age of 28 days) or until the neonates experienced the event of interest (ie, death or censored) were done. Those neonates who were discharged before experiencing the event of interest were followed through phone calls. Only the first episodes of preterm birth in cases where patients were admitted with more than one episode over the 28 days of the initial trial were considered in the current analysis. Six diploma-holder nurses with experience in newborn critical care and three BSc-trained nurses were hired for the data collection and supervision.

    Data quality control

    To preserve consistency, the questionnaire was written in English, translated into Amharic and retranslated into English by linguists. Data collectors and supervisors received training in the study’s objectives, data collection instruments, techniques and procedures before data collection. 32 (5%) preterm neonates who were admitted to the NICU of the Shenen Gibie Hospital underwent a pretest, and the necessary corrections were performed. By closely monitoring the data collectors and supervisors, lead investigators were able to coordinate the overall data collection. Supervisors assessed the questionnaires, and those with missing information were eliminated. The investigators verified the accuracy of the data before they were entered.

    Data processing and analysis

    For cleaning and analysis, data were imported into Epi-Data manager V.4.4 and exported to STATA statistical software V.14. Results are presented as descriptive statistics: frequency, graphs, median and range. To further explain the survival status of preterm newborns, non-parametric estimators including life tables, Kaplan-Meier and log-rank tests were used. For preterm newborns, a semiparametric Cox proportional hazards model was used to determine the important predictors of time to death.

    Variables with a p value of less than 0.25 in the bivariate analysis were entered for the multivariable analysis after each variable underwent Cox proportional hazards regression. The Cox proportional hazards regression model assumption was tested graphically using a log–log plot and statically using the Schoenfeld residual test (global test). Before interpreting the results, multicollinearity between the independent variables was evaluated using variance inflation factor (VIF), and a VIF greater than 10 was used to indicate the presence of multicollinearity. A statistically significant predictor of time to death for preterm neonates was found through the multivariate analysis, with an HR, p value of 0.05 and 95% CI. Finally, the Cox-Snell residual graph and log-likelihood test were used to evaluate the model’s fitness.

    Patient and public involvement

    Patients or the public were not involved in the study design, conduct, reporting or distribution strategies of the research.

    Results

    Neonatal and maternal sociodemographic characteristics

    The analysis comprised of 614 preterm neonates in total, with a response rate of 98.24%. 378 (61.56%) participants were male. The majority of neonates (607, 98.86%) were in the age category of ≤7 days and their median age was 0.68 days. 534 (86.97%) mothers belonged to the age category of 20–34 years, with a median age of 25 years. Two-thirds (66.61%) of the neonates were born to mothers who lived in rural areas. The median follow-up duration for preterm neonates was 5.28 days, and the follow-up period ranged from 1 to 28 days, totalling 3242 person-days (table 1).

    View this table:
    Table 1

    Sociodemographic characteristics of preterm neonates and their mothers admitted to the intensive care unit of Bench Sheko Zone, Keffa Zone and Sheka Zone Hospitals, Southwest Ethiopia, 2021

    Neonatal-related characteristics

    The majority of neonates (586, 95.44%) were born in hospitals, and 470 (76.55%) of them cried immediately after birth. With a mean GA of 32.96, about 368 (59.93%) newborns were born prematurely. With a mean weight of 1560.48, more neonates (54.07%) were born with low birth weight. The 1-minute Apgar scores for 419 of the preterm newborns (71.50%) were less than 7, with an overall median Apgar score of 5.64 and 7.10 for 1-minute and 5-minute scores, respectively. Regarding the type and status of feeding, 506 preterm newborns (82.41%) were given breast milk, and 100 preterm neonates (16.29%) did not start any kind of feeding due to extreme prematurity, surgical and medical complications and being too sick to be fed. The hospital protocol also did not allow the initiation of feeding for neonates born before 32 weeks of gestation. More than half (58.50%) of preterm neonates who had breast milk obtained it after 2 hours from those who had taken it. In terms of weight for GA, 155 neonates (25.24%) were born small for GA, and approximately 180 (29.32%) experienced feeding difficulties (table 2).

    View this table:
    Table 2

    Neonatal-related characteristics of preterm neonates admitted to the intensive care unit of Bench Sheko Zone, Keffa Zone and Sheka Zone Hospitals, Southwest Ethiopia, 2021

    Maternal and obstetrics-related characteristics

    Neonatal births from mothers with a single pregnancy accounted for 470 (76.55%) of neonates. Only 69 (13.69%) of the neonates were delivered from mothers who took corticosteroids throughout pregnancy, out of the 499 (81.27%) neonates who were spontaneously delivered via vaginal birth. Only 64 (10.14.2%) preterm babies were born to mothers who had obstetric problems, and only 46 (7.49%) mothers had chronic disorders (online supplemental file 1).

    Supplemental material

    Medical and surgical preterm-related complications predictors

    Among preterm neonates, nearly two-thirds (394, 64.17%) of neonates had RDS complications and 437 (46.55%) had sepsis issues. Among the total number of babies with RDS diagnosis, 370 neonates (93.91%) underwent resuscitation treatment (online supplemental file 2).

    Institutional and professional characteristics

    In the hospital during neonatal admission, approximately 274 (44.63%) neonates received KMC services, of which 172 (62.77%) KMC services were provided with beds in the NICU (online supplemental file 3).

    Survival status, mortality rate and median survival time among preterm neonates

    200 (32.57%) of the 614 neonates died during the follow-up period, and 414 (67.43%) were censored (of which, 353 (85.23%) were sent home, 39 (9.4%) left against doctors’ orders, 10 (2.4%) were still alive at the end of the study period and the remaining (2.9%) were sent to other institutions). Incidence rate of 61.69 (95% CI: 53.7 to 70.86) deaths per 1000 person-day observations was observed in neonates throughout the course of 3242 person-day observations. Each GA group had a different incidence rate of preterm neonates, with rates for extremely preterm, very preterm and late preterm neonates being 66.67 (95% CI: 16.67 to 266.56), 91.47 (95% CI: 76.37 to 109.5) and 41.62 (95% CI: 33.43 to 51.82), respectively. The median survival time or the survival time at which the cumulative survival function is equal to 0.5 was undetermined. Because the largest observed analysis time was censored, the survivor function did not lead to zero; in this case, the mean is the best estimate of survival time. The mean survival time for preterm neonates was 17.46 (95% CI: 16.34 to 18.98) days.

    Mortality rate and mortality-free survival among preterm neonates

    The study’s Kaplan-Meier survival function estimate revealed that the largest percentage of deaths (81, 49.5%) occurred on the first day of the follow-up period. Additionally, using the Kaplan-Meier estimate of the survivor function, the cumulative survival probabilities at the end of the follow-up periods of 1 day, 7 days and 28 days were 86.81 (95% CI: 83.87 to 89.25), 61.75 (95% CI: 57.10 to 66.05) and 56.92 (95% CI: 51.47 to 61.98), respectively (figure 1).

    Figure 1
    Figure 1

    Overall Kaplan-Meier survival and failure estimates of preterm neonates.

    Comparison of survivorship functions for different categorical variables

    Comparisons of survival time difference between different groups of categorical covariates were done through a Kaplan-Meier survival graph and statistical log-rank test. In this study, in preterm neonates, there were statistically significant differences between groups in place of delivery (Pr>Χ2=0.0001), breathing status (Pr>Χ2=0.0000), feeding status (Pr>Χ2=0.0000), diagnosed RDS (Pr>Χ2=0.0000), asphyxia (Pr>Χ2=0.0129), anaemia (Pr>Χ2=0.0000), 5-minute Apgar score (Pr>Χ2=0.0017) and antenatal care visit (Pr>Χ2=0.0004), as compared with their counterparts (figure 2).

    Figure 2
    Figure 2

    Kaplan-Meier survival curves compare survival time of preterm neonates with the following categories: place of delivery, RDS, hypoglycaemia, breathing status, anaemia, antenatal care visit and mode of delivery. RDS, respiratory distress syndrome.

    Predictors of preterm neonates’ mortality

    The relationship between independent variables and the risk of mortality was analysed using the Cox proportional hazards regression model. In the bivariate analysis, factors such as being male, place of delivery, KMC service, not crying at birth, not initiating feeding for the first time, RDS, asphyxia, hypothermia, anaemia, no antenatal care visit, mode of delivery, no KMC, GA category and 5-minute Apgar score were found to have a p value of <0.25.

    Cox proportional hazards assumption test

    Assumptions of the Cox proportional hazards model were assessed using the Schoenfeld residual/global test, which became non-significant (0.077), indicating that the proportional hazards assumption of Cox proportional hazards regression was met. The multicollinearity of each independent variable was checked using the VIF, and the mean VIF for those variables was 1.35.

    Cox proportional hazards model fitness test

    The fitness of the final model was checked graphically by using the Cox-Snell residual and showed the hazard function followed the 45° line closely, which confirmed that the final model was a good fit (online supplemental file 4).

    Significant variables in the bivariate analysis were included in the multivariable analysis. Finally, sex, KMC service, feeding status, hypothermia, anaemia and GA category were found to be independent predictors of mortality in preterm neonates in the multivariable analysis.

    As a result, there was a 66% (adjusted HR (AHR)=0.66, 95% CI: 0.47 to 0.94) lower risk of death among preterm male neonates compared with preterm female neonates, or a 34% increase in survival time of preterm male neonates compared with preterm female neonates. The risk of death of preterm neonates with a GA lower than 28 weeks was 9.28 times higher (AHR=9.28, 95% CI: 1.78 to 48.42) as compared with those with GA of 28–32 weeks and greater than 32 weeks. The risk of death of preterm neonates who had not been initiated with either formula or breast milk until the time of the first observation was 2.78 times (HR=2.78, 95% CI: 1.8 to 4.28) as compared with those who were initiated breast milk or formula feeding. The risk of death of preterm neonates who were anaemic was 6.2 times (AHR=6.2, 95% CI: 2.34 to 16.43) as compared with that of non-anaemic preterm neonates. Preterm neonates who were not hypothermic had 63% (AHR=0.63, 95% CI: 0.44 to 0.92) lower risk of death as compared with hypothermic preterm neonates. Preterm neonates who got KMC services had 19% (AHR=0.19, 95% CI: 0.12 to 0.29) lower risk of death as compared with preterm neonates who did not get KMC services (table 3).

    View this table:
    Table 3

    Results of the bivariable and multivariable analyses of preterm neonates admitted to the NICU of Bench Sheko Zone, Keffa Zone and Sheka Zone Hospitals, Southwest Ethiopia, 2021

    Discussion

    In this study, the overall incidence rate of mortality was found to be 61.69 (95% CI: 53.7 to 70.86) deaths per 1000 person-day observations, and the highest (81, 49.2%) proportion of deaths was observed on the first day of the follow-up period. The study’s overall findings showed that 32.57% of preterm neonates died, while the mean survival time for preterm neonates was 17.46 (95% CI: 16.24 to 18.89) days.

    The same finding was observed in Jimma University Specialized Hospital where 34.9% of preterm neonates died,23 and in Iran where 28.7% died.24 In contrast, a lower finding was recorded in Gondar University Hospital (25.2%).13 The discrepancy could be attributed to differences in the study design (retrospective vs prospective) that all events are not recorded in the patients’ card (limitation of a retrospective study). Additionally, including a large sample size in our study might have led to a higher record. A retrospective study conducted in Black Lion Hospital showed that the incidence rate was 39.1 per 1000 person-day observations.25 This indicates that the quality of health service provision has a significant role in reducing incidence rates of neonatal mortality. Another study in Northern Ethiopia reported a better mean survival time of the preterm neonates (47.0, 95% CI: 43.1 to 48.9 days)26 compared with this study. This discrepancy might be due to the study setting and the differences in the event of origin and survival time. A lower finding of preterm neonatal deaths that occurred in the first 24 hours of life was observed in the University of Gondar Comprehensive Teaching Hospital (3.29%)13 and Felege Hiwot Referral Hospital (16.6%),19 respectively. This might be due to the variation in NICU services (ie, NICU services at the University of Gondar Comprehensive Teaching Hospital and Felege Hiwot Referral Hospital are organised with personnel and equipment based on newborn conditions (severity) and classified as level 1 (basic), level 2 (specialty) and level 3 (subspecialty)) and study design (retrospective follow-up) that not all deaths might be documented.

    A controversial finding observed in this study was that being male increases a 34% chance of survival (AHR=0.66, 95% CI: 0.47 to 0.94) compared with records observed in Australia (adjusted OR (AOR)=5.7),20 Iran (AOR=1.47)21 and Northeast Brazil (AHR=2.01).14 Possible reasons for this might be the difference in the study population (>61% of participants in this study were male neonates and were only very preterm neonates) and data used (they used secondary data or a retrospective cohort study design).

    A similar research finding in Iran,27 Australia28 and Northeast Brazil21 was observed that neonates with GAs below 28 weeks were substantially more likely to die (AHR=9.28, 95% CI: 1.78 to 48.42) compared with neonates whose GA was greater than 28 weeks. According to this study’s findings, preterm neonates who were not started on breast milk or formula milk had a higher risk of dying than those who were (AHR=2.78, 95% CI: 1.8 to 4.28). This result is comparable with research from Northwest Ethiopia (AHR=0.1021, 95% CI: 0.04480)29 and Uganda (AHR=9.49; 95% CI: 2.84 to 31.68).17

    Additionally, this study showed that anaemia is a substantial predictor of death (AHR=6.2, 95% CI: 2.34 to 16.43), which is comparable with one study done in Northwest Ethiopia (AHR=4.6699, 95% CI: 1.7687 to 12.3297).29 Moreover, the risk of death was 63% (AHR=0.63, 95% CI: 0.44 to 0.92) lower in preterm neonates who were not hypothermic than in those who were. This study and one study completed in Uganda are comparable (AHR=1.98; 95% CI: 1.18 to 3.33).17

    Finally, providing KMC services for preterm neonates can improve their survival time to 81% as compared with that of preterm neonates who did not receive KMC services. This is comparable with a study conducted in Uganda (AHR=9.50; 95% CI: 5.37 to 16.78)17 and Gondar (73%) (AHR=0.25, 95% CI: 0.13, 0.58).13

    Since this study was conducted in a multicentre setting, it increases the generalisability of the findings to the entire population. Moreover, this study was prospective; it helps to address probable service-related predictors and control missing data. As the study was conducted in a healthcare setting, those preterm neonates who were born at home and were not admitted to the hospital were not incorporated. Further, this study had a limitation in physician diagnosis subjectivity due to the fact that we did not use consistent diagnostic criteria for each participant, since we already used what they had been diagnosed using history, physical examination and laboratory diagnostic criteria. In addition, this study may lead to missing associated factors after leaving the hospital since this study uses phone calls to follow up for those neonates who were discharged before experiencing the event of interest.

    Conclusion and recommendation

    This study included a significant proportion of premature newborn deaths. Poor health indicators in preterm neonates, such as delayed initiation of breast feeding and formula feeding, lack of access to a KMC service, home birth, GA of 28 weeks and low initial temperature, have been proven to be predictors of mortality in preterm neonates. Similarly, preterm infants with female sex and anaemia were less likely to survive. Therefore, healthcare workers should encourage KMC services and breastfeeding initiation and avoid bleeding complications to prevent neonates from being anaemic and to increase their chances of survival. Future studies should assess the level of awareness, treatment and control of these risk factors.

    Data availability statement

    Data are available upon reasonable request. all data are incorporated within the manuscript

    Ethics statements

    Patient consent for publication

    Ethics approval

    A formal letter of collaboration was subsequently written for each hospital after receiving ethical approval from the Mizan-Tepi University Ethical Review Board (ref. no: R/C/S/D/0023/2012). The mothers of the research participants were informed of the goals and aims of the study, and their input was essential to produce accurate and beneficial data. Each mother was given an explanation of the study’s goals, selection criteria, confidentiality and advantages before providing oral informed consent. To maintain the confidentiality of the participants, all permission and data collection processes were conducted in a private setting. The participants were advised by the data collectors that they could pause, stop or end the survey at any moment.

    Acknowledgments

    We would like to express our deepest gratitude to Mizan-Tepi University College of Health Science for giving us the opportunity to conduct this study and we also thank the NICU staff of each hospital for providing the necessary information that we required. We extend our special appreciation to the study participants, data collectors and supervisors.

    References

    1. ↵
      2. Soon BT
      . The Global Action Report on Preterm Birth. Geneva: World Health Organization, 2012:2.
    2. ↵
      2. Offiah I ,
      3. O’Donoghue K ,
      4. Kenny L
      . Clinical Risk Factors for Preterm Birth. Preterm Birth: Mother and Child. Rijeka, Croatia: InTech, 2012:73–94.
    3. ↵
      2. Howson C ,
      3. Kinney M ,
      4. Lawn J
      . Born Too Soon: The Global Action Report on Preterm Birth March of Dimes, PMNCH. Save the Children. Geneva: WHO World Health Organization, 2012.
    4. ↵
      2. Harrison MS ,
      3. Goldenberg RL
      . Global burden of Prematurity. Semin Fetal Neonatal Med 2016;21:74–9. doi:10.1016/j.siny.2015.12.007
      OpenUrlCrossRefPubMed
    5. ↵
      2. Liu L ,
      3. Johnson HL ,
      4. Cousens S , et al
      . Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. The Lancet 2012;379:2151–61. doi:10.1016/S0140-6736(12)60560-1
      OpenUrlCrossRef
    6. ↵
      2. Liu L ,
      3. Oza S ,
      4. Hogan D , et al
      . Global, regional, and national causes of Under-5 mortality in 2000–15: an updated systematic analysis with implications for the sustainable development goals. The Lancet 2016;388:3027–35. doi:10.1016/S0140-6736(16)31593-8
      OpenUrlCrossRef
    7. ↵
      2. Abdel Razeq NM ,
      3. Khader YS ,
      4. Batieha AM
      . The incidence, risk factors, and mortality of Preterm neonates: A prospective study from Jordan (2012-2013). Tjod 2017;14:28–36. doi:10.4274/tjod.62582
      OpenUrl
    8. ↵
      2. Liu L ,
      3. Oza S ,
      4. Hogan D , et al
      . Global, regional, and national causes of child mortality in 2000–13, with projections to inform Post-2015 priorities: an updated systematic analysis. The Lancet 2015;385:430–40. doi:10.1016/S0140-6736(14)61698-6
      OpenUrlCrossRef
    9. ↵
      2. Onankpa B ,
      3. Isezuo K
      . Pattern of Preterm delivery and their outcome in a tertiary hospital. Int J Health Sci Res 2014;4:59–65.
      OpenUrl
    10. ↵
      2. Mengesha HG ,
      3. Wuneh AD ,
      4. Lerebo WT , et al
      . Survival of neonates and predictors of their mortality in Tigray region, northern Ethiopia: prospective cohort study. BMC Pregnancy Childbirth 2016;16:202. doi:10.1186/s12884-016-0994-9
    11. ↵
      2. Mekonnen T ,
      3. Tenu T ,
      4. Aklilu T , et al
      . Assessment of neonatal death and causes among admitted neonates in neonatal intensive care unit of Mizan Tepi University teaching hospital, bench Maji zone, South-West Ethiopia, 2018. Clinics Mother Child Health 2018;15:2. doi:10.4172/2090-7214.1000305
      OpenUrl
    12. ↵
      2. Lawn JE ,
      3. Blencowe H ,
      4. Oza S , et al
      . Every newborn: progress, priorities, and potential beyond survival. The Lancet 2014;384:189–205. doi:10.1016/S0140-6736(14)60496-7
      OpenUrlCrossRefPubMed
    13. ↵
      2. Yismaw AE ,
      3. Gelagay AA ,
      4. Sisay MM
      . Survival and predictors among Preterm neonates admitted at University of Gondar comprehensive specialized hospital neonatal intensive care unit, Northwest Ethiopia. Ital J Pediatr 2019;45:4. doi:10.1186/s13052-018-0597-3
    14. ↵
      2. Lawn JE ,
      3. Mwansa-Kambafwile J ,
      4. Horta BL , et al
      . Kangaroo mother Care’To prevent neonatal deaths due to Preterm birth complications. Int J Epidemiol 2010;39 Suppl 1:i144–54. doi:10.1093/ije/dyq031
      OpenUrlPubMed
    15. ↵
      2. Kunle-Olowu OE ,
      3. Peterside O ,
      4. Adeyemi OO
      . Prevalence and outcome of Preterm admissions at the neonatal unit of a tertiary health centre in Southern Nigeria. OJPed 2014;04:67–75. doi:10.4236/ojped.2014.41009
      OpenUrl
    16. ↵
      2. Parappil H ,
      3. Rahman S ,
      4. Salama H , et al
      . Outcomes of 28+ 1 to 32+ 0 weeks gestation babies in the state of Qatar: finding facility-based cost effective options for improving the survival of Preterm neonates in low income countries. Int J Environ Res Public Health 2010;7:2526–42. doi:10.3390/ijerph7062526
      OpenUrlPubMed
    17. ↵
      2. Tibaijuka L ,
      3. Bawakanya SM ,
      4. Owaraganise A , et al
      . Incidence and predictors of Preterm neonatal mortality at Mbarara regional referral hospital in South Western Uganda. PLoS One 2021;16:e0259310. doi:10.1371/journal.pone.0259310
    18. ↵
      2. Opio C ,
      3. Malumba R ,
      4. Kagaayi J , et al
      . Survival time and its predictors among Preterms in the neonatal period post-discharge in Busoga region-Uganda June–July 2017. Journal of Interventional Epidemiology and Public Health 2019;2:2. doi:10.37432/JIEPH.2019.2.2.19
      OpenUrl
    19. ↵
      2. Tamene A ,
      3. Abeje G ,
      4. Addis Z
      . Survival and associated factors of mortality of Preterm neonates admitted to Felege Hiwot specialized hospital, Bahir Dar, Ethiopia. SAGE Open Med 2020;8:2050312120953646. doi:10.1177/2050312120953646
    20. ↵
      2. Schindler T ,
      3. Koller-Smith L ,
      4. Lui K , et al
      . Causes of death in very Preterm infants cared for in neonatal intensive care units: a population-based retrospective cohort study. BMC Pediatr 2017;17:59. doi:10.1186/s12887-017-0810-3
    21. ↵
      2. Basiri B ,
      3. Esna Ashari F ,
      4. Shokouhi M , et al
      . Neonatal mortality and its main determinants in premature infants hospitalized in neonatal intensive care unit in Fatemieh hospital, Hamadan, Iran. J Compr Ped 2015;6. doi:10.17795/ijcp-26965
    22. ↵
      2. Gargari SS
      . Survival and risk factors of extremely Preterm babies (< 28 weeks) in the three Iranian hospitals. Acta Med Iran 2018;181–8.
    23. ↵
      2. Wesenu M ,
      3. Kulkarni S ,
      4. Tilahun T
      . Modeling determinants of time-to-death in premature infants admitted to neonatal intensive care unit in Jimma University specialized hospital. Ann Data Sci 2017;4:361–81. doi:10.1007/s40745-017-0107-2
      OpenUrl
    24. ↵
      2. Ghorbani F ,
      3. Saeed D ,
      4. Mohammad H , et al
      . Survival of premature and low birth weight infants: A multicenter, prospective, cohort study in Iran. Iranian Journal of Neonatology 2017;8.
    25. ↵
      2. Asmare Y
      . Survival status and Predictor of mortality among premature neonate that was admitted to neonatal intensive care unit from 2013-2017 at Tikur Anbessa hospital, Addis Ababa Ethiopia: conference presentation. Addis Ababa Univ 2018;8:4172.
      OpenUrl
    26. ↵
      2. Gebreheat G ,
      3. Teame H
      . Survival and mortality of Preterm neonates in a neonatal intensive care unit in northern Ethiopia: a retrospective cohort study. Sci Rep 2022;12:600. doi:10.1038/s41598-021-04521-z
    27. ↵
      2. Abdel-Latif ME ,
      3. Bajuk B ,
      4. Oei J , et al
      . Does rural or urban residence make a difference to neonatal outcome in premature birth? A regional study in Australia. Arch Dis Child Fetal Neonatal Ed 2006;91:F251–6. doi:10.1136/adc.2005.090670
      OpenUrlAbstract/FREE Full Text
    28. ↵
      2. Castro ECM de ,
      3. Leite ÁJM ,
      4. Almeida MFB de , et al
      . Perinatal factors associated with early neonatal deaths in very low birth weight Preterm infants in northeast Brazil. BMC Pediatr 2014;14:312. doi:10.1186/s12887-014-0312-5
    29. ↵
      2. Yehuala S ,
      3. Teka Z
      . Survival analysis of premature infants admitted to neonatal INT Ensive care unit (NICU) in Northwest Ethiopia using semi-parametric FR Ailty model. Journal of Biometrics & Biostatistics 2015;6:1.
      OpenUrl

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors EM conceptualised the study, wrote the proposal, and performed data analysis and manuscript preparation. Beside this EM take the overall responsibility as principal investigator(PI) for the work. YDG performed data analysis and wrote the results. EA supervised subsequent drafts of the paper. ATS supervised the data analysis. All authors contributed to the article and approved the submitted version.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.