{
  "abstract": "Objectives This study aimed to develop and internally validate a neonatal mortality risk prediction model for low birthweight (LBW) neonates in public hospitals of North-West Ethiopia.Design An institution-based prospective follow-up study.Setting The study was conducted in seven hospitals, comprising three general hospitals and four primary hospitals, between 14 February and 31 October 2025. The general hospitals, Injibara, Finote Selam and Pawi, provide level II neonatal intensive care services, whereas the primary hospitals, Dangila, Chagni, Jawi and Gimja Bet, provide level I newborn care services. These hospitals are situated in the Awi, West Gojjam and Metekel zones of the Amhara and Benishangul Gumuz regions in north-western Ethiopia.Participants 955 LBW neonates were enrolled in the study. Neonates born outside health facilities were excluded if their birth weight was unknown, particularly those delivered by traditional birth attendants, as were neonates whose first recorded weight was obtained more than 24 hours after birth. Participants were selected using consecutive sampling and data were collected electronically using the KoboCollect mobile application.Outcome measures Data were analysed using R software V.4.4.3. Least Absolute Shrinkage and Selection Operator regression and multivariable logistic regression were used to identify predictors and develop a simplified risk score and nomogram. Model performance was evaluated in terms of discrimination and calibration, while internal validation was performed using bootstrapping. The clinical utility of the model was assessed using decision curve analysis (DCA).Results The final model included nine predictors: sepsis, perinatal asphyxia, respiratory distress syndrome, absence of exclusive breastfeeding, absence of Kangaroo Mother Care, gestational ages of 28 weeks to <32 weeks and 32 weeks to <37 weeks, and birth weights of <1000 g and 1000–1499 g. The original model demonstrated good discrimination, with an area under the curve of 84.2% (95% CI 81.5% to 86.9%). After internal validation, the simplified risk score and nomogram achieved areas under the curve of 83.85% and 83.5%, respectively. The model showed excellent calibration before and after validation. DCA indicated that the model provides meaningful clinical benefit.Conclusions The nomogram and simplified risk score demonstrated excellent calibration and good discriminatory performance. Their application in clinical settings may improve early intervention and potentially reduce neonatal mortality in Ethiopia by enabling rapid, individualised mortality-risk prediction among LBW neonates. However, external validation in different settings is required.",
  "authors": [
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Daniel Mulat Eshetu"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Assefa Andargie Kassa"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Wolde Melese Ayele"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Mulatu Kassahun"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Shashitu Tadele Zeleke"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Betelhem Mekuriaw Anagaw"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Mikias Getahun Molla"
    },
    {
      "affiliations": [
        "Department of Internal Medicine, School of Medicine, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Kalaab Esubalew Sharew"
    },
    {
      "affiliations": [
        "Biomedical Sciences Education and Service Directorate, College of Health Sciences, Debre Tabor University, Debre Tabor, Ethiopia"
      ],
      "name": "Birhanu Kassie Abeje"
    },
    {
      "affiliations": [
        "Department of Public Health, College of Medicine and Health Sciences, Injibara University, Injibara, Ethiopia"
      ],
      "name": "Chalachew Abiyu Ayalew"
    }
  ],
  "title": "Development and internal validation of a neonatal mortality risk prediction model for low birthweight neonates in public hospitals, North-West Ethiopia: a prospective follow-up study",
  "uid": "f9b478ba-edec-5a0f-ae0e-1e59549c26b7"
}