Study design. From January 1, 2016, to December 31, 2020, a prospective observational study was conducted to explore the trends in neonatal mortality and to unravel the distribution characteristics of the causes underlying neonatal death in Shaanxi province. To ensure a geographically balanced representation of neonates across Shaanxi Province, we employed stratified and cluster sampling methods. In general, six neonatal medical centers were selected, including one provincial and four municipal rescue transit centers dedicated to critically ill neonates. These centers predominantly serve approximately half of the province's areas, covering 5 cities.
In our study, neonates were admitted from the outpatient, neonatology department, emergency department, delivery or operating room in own medical centers. We also included these neonates, who came from other hospitals and were transported to the study medical centers. During the study period, a total of 220,488 neonates were born in these centers, representing 12% of all live births across the entire province10. Among these, 73,967 neonates were admitted for care, and there were 420 neonatal deaths (Fig. 1).
Data collection. In this study, we included neonates who were hospitalized and died within 28 days of birth. Exclusion criteria were neonates born with a gestational age of less than 24 weeks or those lacking comprehensive medical records. Our definition of neonatal mortality encompassed not only deaths occurring within in medical centers, neonates who hospitalized in our medical centers, but for various reasons, their parents refused any further treatments and requested discharge from the medical centers, and died in 24 hours after discontinuation of treatment also included. The following neonatal information was extracted from medical records: gender, gestational age, birth date, birth weight, APGAR scores at 1 minute, APGAR scores at 5 minutes, and other neonatal information regarding birth, medication and information on deaths. Additionally, baseline data on maternal antenatal and perinatal history were collected during the initial antenatal care check, recorded by professional physicians.
Variable definition. The determination of the underlying causes of neonatal death was based on hospital records. For cases subject to review by provincial experts, the final diagnosis was established based on the experts’ assessments. All causes of death were diagnosed based on the diagnostic criteria of the Practical Neonatology11 and were coded following the International Classification of Diseases, Tenth Revision12, by pediatricians. In our study, we combined death causes except neonatal respiratory distress syndrome (NRDS), birth asphyxia, neonatal septicemia, and congenital malformation as ‘others’ for analysis. Notably, NRDS, birth asphyxia, neonatal septicemia, and congenital malformation accounted for 76.4% (n = 321) of all neonatal deaths.
The gestational age at birth was estimated by the mother’s last menstrual period (LMP), calculating completed days from the first day of the LMP. Natural conception was defined as conception occurring without the aid of assisted reproductive technologies. The method of childbirth was categorized as either 'vaginal' or 'cesarean' depending on the delivery mode. Single/multiple birth were determined on the basis of the number of fetuses delivered. Preterm was defined as a gestational age of less than 37 completed weeks. LBW was defined as a birth weight of newborn less than 2,500 grams. Small for gestational age (SGA) was determined when a newborn’s weight fell below the 10th percentile, according to Chinese growth standards for various gestational ages13. Maternal disease during pregnancy encompassed conditions diagnosed by physicians, including hypertensive disorders complicating pregnancy, gestational diabetes mellitus, gestational hypothyroidism, or other gestational diseases.
Data analysis. The background characteristics of the mothers and neonates were described by means and standard deviations for continuous variables, and percentages for categorical variables. The neonatal mortality rate was calculated as the number of neonatal deaths divided by the total number of neonates hospitalized during the same period in the medical centers. Linear trend test was used to explore the trend of neonatal mortality rate. Multivariable logistic regression models were used to analyze the distribution of the causes of neonatal deaths, accounted for a variety of variables such as maternal age, maternal diseases during pregnancy, methods of conception, parity, number of fetuses, modes of delivery, preterm, LBW, SGA, and neonatal gender. Statistical analyses were conducted with SPSS version 22.0 (SPSS Inc., Chicago, IL, USA). A two-tailed P < 0.05 was considered statistically significant. All methods were performed in accordance with the relevant guidelines and regulations.
Ethics approval.
The study was approved by the Ethical Committee of Northwest Women’s and Children’s Hospital (protocol no. 2022002). Written informed consent was obtained the legal guardians of all included neonates. The research was performed in accordance with relevant national and international guidelines/regulations.