A retrospective follow-up study was conducted among preterm neonates admitted to NICU at JUMC between 1 January 2017, and 30 December 2019. JUMC is found in Jimma town 352 km away from Addis Ababa, the capital city of Ethiopia, in the southwestern part of the country. JUMC currently provides a range of services for approximately 15 million people. The NICU unit is one of the intensive care unit services currently in operation at the hospital which has 26 neonatal and 16 kangaroo-mother care beds. The unit also has 20 radiant warmers, four continuous positive airway pressure (CPAP), six photo-therapy machines, oxygen concentrator machines, pulse oximetry, a glucometer, and neonatal resuscitation equipment. On average, nearly 1,500 neonates are admitted annually to NICU of JUMC. The functional capability of JUMC is level three NICU [25]. The data collection period was from March 11 to 20 April 2020.

The source population for this study was all preterm neonates admitted to the NICU of JUMC from 1 January 2017, to 30 December 2019. All randomly selected preterm neonates admitted to the NICU of JUMC from 1 January 2017, to 30 December 2019, and fulfilling the eligibility criteria were the study population. All alive-born preterm neonates at admission who were registered on the neonatal registry book from 1 January 2017, to 30 December 2019, in the NICU of JUMC were included in the study. However, preterm neonates with incomplete information on medical records regarding outcome status, a time when neonates were admitted to NICU, and a time when death or censoring occurred were excluded.

The sample size was determined for survival analysis by considering preterm neonates who have jaundice using STATA Version 15 statistical software based on the following assumptions: 5% level of significance (α) (two-sided), 80% power, adjusted hazard ratio of 1.62 for preterm neonates who have jaundice, overall probability of preterm neonatal death (d) of 0.288 [15], and 0.5 variabilities of covariates of interest. It was assumed that no subjects were anticipated to withdraw from the follow-up, and a 10% contingency was added for incomplete records. Hence, the total sample size for this study was 516.

The medical registration number (MRN) of preterm neonates over three 3-year period from 1 January 2017, to 30 December 2019, was taken from the NICU logbook to create a sampling frame. A total of 957 preterm neonates were admitted to NICU. A computer-generated simple random sampling technique was employed to select 516 participants for the study as follows: The sampling frame that was created using the MRN was entered into SPSS version 25 software. Then, a 516 sample was selected randomly using the SPSS select case procedure. Medical records of preterm neonates attached to selected MRNs were reviewed, and those records that met eligibility criteria were included in the analysis.

Time to death of preterm neonates was the dependent variable (death was an event and coded “1,” and censored observation coded “0”). Socio-demographic related variables such as sex of neonate, neonatal age at admission, maternal age, and residence; maternal medical condition-related variables like maternal febrile illness/disease, anemia, diabetes mellitus, and human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS); obstetric and gynecologic related variables like gravidity, antenatal care visit, birth type, history of bad obstetric and/or gynecologic outcome, mode of delivery, presentation, place of delivery, antenatal steroid use, and obstetric complications; and preterm neonates related variables such as gestational age at birth, birth weight, weight class for gestational age, fifth minute APGAR score, kangaroo-mother care, initiation of breastfeeding, method of feeding, congenital malformation, temperature, respiratory distress syndrome, neonatal sepsis, perinatal asphyxia, hypoglycemia, anemia, and jaundice were independent variables.

Data were collected from preterm neonatal medical records and registers by three trained bachelor’s degree holder midwives and supervised by one bachelor’s degree holder senior nurse. A data collection checklist adapted from the Global Neonatal Database data collection form for the Ethiopian Neonatal Network was used to collect the data [26]. Modifications were made to the checklist based on NICU registration format, and through reviewing relevant literature. The starting point for follow-up was the first NICU admission date and followed until the last neonatal period (28th days of life), which was the endpoint of the study.

➢

Survival status: An outcome of the neonate during follow-up from the medical records and is considered as “death” if neonate died during follow-up, as “lost to follow-up” if the mother or caregiver was not available and unable to reach their address. It was considered as “Withdrawal” if the mother refused the follow-up due to inconvenience, as “refereed” if the neonate was referred to other institutions for better management, and “alive” if the preterm neonate survival was assured at the last follow-up period [27].

Data quality was assured by the careful designing of the data collection checklist, recruiting data collectors, and supervisor who have previous experience. A preliminary chart review was done on 26 randomly selected records (5% of the sample size) before the commencement of the actual study, and relevant clarifications and amendments were taken on the checklist. Training for 2 days was given on principles of research ethics, data collection checklist, and procedures for data collectors and a supervisor. Data collectors were supervised closely by the supervisor daily throughout the data collection period.

Data were cleaned, coded, and entered into Epi-Data version 3.1, and analysis was done using STATA version 15.0. Descriptive statistics such as frequencies, percentages, summary measures, and rates were computed to describe categorical and continuous variables as supposed necessary. The incidence rate of neonatal mortality was computed by dividing the number of preterm neonates who died during the follow-up period by the total neonate-days at risk of observation. The Kaplan-Meier (KM) method was used to estimate median survival time and compare survival experience between categories of variables. Log-rank test was used to compare statistically significant differences in survival experience among groups.

The Cox proportional hazard model was used to identify predictors of time to death. A bivariable cox-proportional hazard model was fitted first, and variables with a p-value <0.25 in this analysis entered into the multivariable cox-proportional hazard model. To identify independent predictors of time to death, a stepwise backward likelihood ratio method was used to fit a multivariable cox-proportional hazard model. An adjusted hazard ratio with a 95% confidence interval was computed to determine the strength of the association. Variables with a p-value <0.05 in the final model were considered significant predictors of the time to death of preterm neonates.

The proportional hazard assumption was checked by the Schoenfeld residual test and was satisfied (Global test X² = 5.13, p-value = 0.92). Multicollinearity was checked by looking at the variance inflation factor (VIF) and the highest observed VIF value was 2.06, indicating that there was no multicollinearity threat. The goodness of model fitness was evaluated by using the Cox-Snell residual test. In this study, the Nelson-Aalen cumulative hazard function follows the 45° diagonal line very closely, indicating that it almost has an exponential distribution with a hazard rate of one. Hence, for the residual test, it was possible to conclude that the final model fits the data very well (Figure 1).

Near to nine-tenth of neonates, 433 (85.7%), had less than 24 h of age at admission and more than half, 279 (55.2%), of them were males. The median age of the mother was 27 years (an interquartile range (IQR) of 8 years). Most mothers of the neonates, 398 (78.8%), were between the ages of 20 and 34. Nearly two-thirds, 339 (67.1%), of preterm neonates were rural residents (Table 1).

Of the participants, 66 (13.1%) mothers had known or been diagnosed with a medical disease, and more than nine-tenths of the mothers, 467 (92.5%), had antenatal care visits during the current pregnancy. Almost a quarter, 124 (24.6%), of the mothers had used antenatal steroids. Nearly three-fourths, 372 (73.7%), of the mothers had spontaneous onset of labor, and the majority of the delivery, 462 (91.5%), had a cephalic presentation. Almost half, 250 (49.5%), of the mothers had an obstetric complication (Table 2).

Almost four-fifths of the neonates, 399 (79%), were moderate preterm, and more than two-thirds of neonates, 349 (69%), had low birth weight. Out of the cohort, 454 (89.9%) had not initiated breastfeeding within 1 hour of birth. More than three-fourths of neonates, 397 (78.6%), were diagnosed with hypothermia followed by respiratory distress syndrome 295 (58.4%), and hypoglycemia 167 (33.1%). Kangaroo-mother care (KMC) was provided to nearly one in every ten neonates (30.7%). More than three-fifths of neonates, 59.6%, received nasal CPAP (Table 3).

During the follow-up, the cumulative incidence of preterm neonatal death was 127 (25.1%). Of all deaths, 15.7% died in the first 24 h of life, and 81.1% of deaths occurred within 7 days of life. Out of the cohort, 352 (69.7%) improved and were discharged to home, 15 (3.0%) lost to follow-up, 6 (1.2%) were referred to other hospitals, and the remaining 5 (1.0%) were withdrawn from the follow-up.

A cohort contributed a total of 4,386 neonate days at risk of observation. The overall neonatal mortality rate (incidence density) was 28.9 deaths per 1,000 neonate-days (95% CI: 24.33, 34.46). The neonatal mortality rate (NMR) was 67.3 deaths per 1,000 neonate-days in the first 24 h of life (95% CI: 48.11, 94.23). Early NMR (death within 7 days of life) was 40 deaths per 1,000 neonate-days (95% CI: 33.08, 48.33); however, the late NMR was 11.7 deaths per 1,000 neonate-days (95% CI: 7.55, 18.13).

Preterm neonates were followed for different periods: a minimum of 1 day and a maximum of 28 days. The overall median length of follow-up was 7 (IQR = 8) days. The cumulative survival probability at the end of the follow-up was 54.94% (95% CI: 41.83, 66.27). The cumulative probability of survival at the end of the first, seventh, 14th, and 21st days was 93.27% (95% CI: 90.71, 95.14), 76.89% (95% CI: 72.73, 80.51), 71.8% (95% CI: 66.79, 76.19), and 66.96% (60.54, 72.58), respectively. The overall mean survival time was 20.42 neonate days (95% CI: 19.27, 21.56).

Preterm neonates born from mothers who used antenatal steroids had higher survival experiences compared to their counterparts (chi-square = 5.17, p-value = 0.023). Likewise, preterm neonates who received KMC had a higher survival experience than neonates who didn’t receive KMC (chi-square = 14.18, p-value = 0.0002). However, preterm neonates born from mothers with obstetric complications had lower survival experience than their counterparts (chi-square = 11.71, p-value = 0.001). Neonates having respiratory distress syndrome (RDS) had lower survival experiences than neonates without RDS (chi-square = 11.14, p-value = 0.001). Preterm neonates with neonatal sepsis had lower survival experiences than their complements (chi-square = 7.55, p-value = 0.006). Neonates with perinatal asphyxia (PNA) had lower survival experiences than their counterparts (chi-square = 7.51, p-value = 0.003) (Figure 2).

In the multivariable cox-regression model; antenatal steroid use, obstetric complication during the current pregnancy, an increment in gestational age at birth, receiving KMC, having RDS, neonatal sepsis, and PNA were found to be predictors for time to death of preterm neonates at p-value <0.05.

Preterm neonates born from mothers who used antenatal steroids during current pregnancy had 45% fewer hazard of death compared to neonates born from mothers who didn’t use antenatal steroids (AHR = 0.55; 95% CI:0.34, 0.90). Preterm neonates born from mothers with an obstetric complication had a 1.84 times higher hazard of death compared to those who were born from mothers without obstetric complication (AHR = 1.84; 95% CI: 1.20, 2.82).

As the gestational age of preterm neonates at birth increases by 1 week, the hazard of death decreases by 19% (AHR = 0.81; 95% CI: 0.75, 0.87). Preterm neonates who had RDS had 1.52 times more hazard of death than those without RDS (AHR = 1.52; 95% CI:1.01, 2.29). Preterm neonates who had neonatal sepsis had about 1.71 times greater hazard of death than neonates without neonatal sepsis (AHR = 1.71; 95% CI: 1.18, 2.49). Preterm neonates who had PNA had 2.44 times more hazard of death compared to those neonates without PNA (AHR = 2.44; 95% CI: 1.33, 4.49). Preterm neonates who received KMC had 52% lesser hazard of death as compared to their counterparts (AHR = 0.48; 95% CI: 0.30, 0.77) (Table 4).

The incidence rate of preterm neonatal mortality was found high. Most preterm neonatal mortality occurs in the early phase of the neonatal period, which requires due attention to meet the national newborn and child survival and SDG-3 in Ethiopia. Obstetric complications, respiratory distress syndrome, neonatal sepsis, and perinatal asphyxia were found to be predictors of preterm neonatal mortality. Whereas, antenatal steroid exposure, an increment in gestational age at birth, and receiving kangaroo-mother care were preventive predictors for preterm neonatal mortality. Hence, special emphasis and close follow-up are highly warranted, especially during the early neonatal period. It is better to strengthen obstetrics and use of antenatal steroids for women having an imminent preterm delivery. Early detection and management of obstetric as well as preterm neonatal complications is highly demanding. Encouraging and supporting mothers to practice kangaroo-mother care, and ensuring a continuum of care are also crucial to enhance preterm neonatal survival.