Development and validation of a simplified score to predict neonatal mortality risk among neonates weighing 2000 g or less (NMR-2000): an analysis using data from the UK and The Gambia

Geographical validation of the Smart Triage Model by age group

Infectious diseases in neonates account for half of the under-five mortality in low- and middle-income countries. Data-driven algorithms such as clinical prediction models can be used to efficiently detect critically ill children in order to optimize care and reduce mortality. Thus far, only a handful of prediction models have been externally validated and are limited to neonatal in-hospital mortality. The aim of this study is to externally validate a previously derived clinical prediction model (Smart Triage) using a combined prospective baseline cohort from Uganda and Kenya with a composite endpoint of hospital admission, mortality, and readmission. We evaluated model discrimination using area under the receiver-operator curve (AUROC) and visualized calibration plots with age subsets (< 30 days, ≤ 2 months, ≤ 6 months, and < 5 years). Due to reduced performance in neonates (< 1 month), we re-estimated the intercept and coefficients and selected new thresholds to maximize sensitivity and specificity. 11595 participants under the age of five (under-5) were included in the analysis. The proportion with an endpoint ranged from 8.9% in all children under-5 (including neonates) to 26% in the neonatal subset alone. The model achieved good discrimination for children under-5 with AUROC of 0.81 (95% CI: 0.79-0.82) but poor discrimination for neonates with AUROC of 0.62 (95% CI: 0.55-0.70). Sensitivity at the low-risk thresholds (CI) were 85% (83%-87%) and 68% (58%-76%) for children under-5 and neonates, respectively. After model revision for neonates, we achieved an AUROC of 0.83 (95% CI: 0.79-0.87) with 13% and 41% as the low- and high-risk thresholds, respectively. The updated Smart Triage performs well in its predictive ability across different age groups and can be incorporated into current triage guidelines at local healthcare facilities. Additional validation of the model is indicated, especially for the neonatal model.

Effectiveness of kangaroo mother care before clinical stabilisation versus standard care among neonates at five hospitals in Uganda (OMWaNA): a parallel-group, individually randomised controlled trial and economic evaluation

BACKGROUND

Preterm birth is the leading cause of death in children younger than 5 years worldwide. WHO recommends kangaroo mother care (KMC); however, its effects on mortality in sub-Saharan Africa and its relative costs remain unclear. We aimed to compare the effectiveness, safety, costs, and cost-effectiveness of KMC initiated before clinical stabilisation versus standard care in neonates weighing up to 2000 g.

METHODS

We conducted a parallel-group, individually randomised controlled trial in five hospitals across Uganda. Singleton or twin neonates aged younger than 48 h weighing 700-2000 g without life-threatening clinical instability were eligible for inclusion. We randomly assigned (1:1) neonates to either KMC initiated before stabilisation (intervention group) or standard care (control group) via a computer-generated random allocation sequence with permuted blocks of varying sizes, stratified by birthweight and recruitment site. Parents, caregivers, and health-care workers were unmasked to treatment allocation; however, the independent statistician who conducted the analyses was masked. After randomisation, neonates in the intervention group were placed prone and skin-to-skin on the caregiver's chest, secured with a KMC wrap. Neonates in the control group were cared for in an incubator or radiant heater, as per hospital practice; KMC was not initiated until stability criteria were met. The primary outcome was all-cause neonatal mortality at 7 days, analysed by intention to treat. The economic evaluation assessed incremental costs and cost-effectiveness from a disaggregated societal perspective. This trial is registered with ClinicalTrials.gov, NCT02811432.

FINDINGS

Between Oct 9, 2019, and July 31, 2022, 2221 neonates were randomly assigned: 1110 (50·0%) neonates to the intervention group and 1111 (50·0%) neonates to the control group. From randomisation to age 7 days, 81 (7·5%) of 1083 neonates in the intervention group and 83 (7·5%) of 1102 neonates in the control group died (adjusted relative risk [RR] 0·97 [95% CI 0·74-1·28]; p=0·85). From randomisation to 28 days, 119 (11·3%) of 1051 neonates in the intervention group and 134 (12·8%) of 1049 neonates in the control group died (RR 0·88 [0·71-1·09]; p=0·23). Even if policy makers place no value on averting neonatal deaths, the intervention would have 97% probability from the provider perspective and 84% probability from the societal perspective of being more cost-effective than standard care.

INTERPRETATION

KMC initiated before stabilisation did not reduce early neonatal mortality; however, it was cost-effective from the societal and provider perspectives compared with standard care. Additional investment in neonatal care is needed for increased impact, particularly in sub-Saharan Africa.

FUNDING

Joint Global Health Trials scheme of the Department of Health and Social Care, Foreign, Commonwealth and Development Office, UKRI Medical Research Council, and Wellcome Trust; Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Effects of implementation of a care bundle on rates of necrotising enterocolitis and own mother's milk feeding in the East Midlands: protocol for a mixed methods impact and process evaluation study

INTRODUCTION

Prevention of necrotising enterocolitis (NEC) is vital for improving neonatal outcomes. Feeding own mother's milk helps prevent NEC. Rates of own mother's milk feeding in the East Midlands are lower than the national average and the incidence of NEC is higher. The East Midlands Neonatal Operational Delivery Network (EMNODN) has created a care bundle to improve these in babies born at <32 weeks' gestation, the group at the highest risk of NEC. The bundle was introduced in September 2022 and embedded by December 2022. We will evaluate its effectiveness and conduct a process evaluation to understand barriers and facilitators to implementation.

METHODS AND ANALYSIS

We will conduct a retrospective cohort study (workstream 1) using data from the National Neonatal Research Database (NNRD). We will identify infants receiving any own mother's milk on day 14 and at discharge, and cases of severe NEC. We will aggregate outcomes by birth month and use interrupted time series analysis to estimate an incidence rate ratio for changes after the care bundle was embedded, relative to pre-implementation. We will model data from all other NNRD units and assess whether there are any concurrent changes to exclude confounding due to other events.We will apply the RE-AIM framework (workstream 2), supplemented by the Consolidated Framework for Implementation Research and Framework for Implementation Fidelity, to conduct a mixed methods evaluation in EMNODN units. We will triangulate data from several sources, including questionnaires and semistructured interviews with parents and healthcare professionals, and data from patient records.

ETHICS AND DISSEMINATION

The study has approval from the South East Scotland Research Ethics Committee 01 and the Health Research Authority and Health and Care Research Wales (IRAS 323099). Results will be disseminated via scientific journals and conferences, to neonatal service commissioners and through public-facing infographics.

TRIAL REGISTRATION NUMBER

NCT05934123.

Nomogram to predict risk of neonatal mortality among preterm neonates admitted with sepsis at University of Gondar Comprehensive Specialized Hospital: risk prediction model development and validation

BACKGROUND

Mortality in premature neonates is a global public health problem. In developing countries, nearly 50% of preterm births ends with death. Sepsis is one of the major causes of death in preterm neonates. Risk prediction model for mortality in preterm septic neonates helps for directing the decision making process made by clinicians.

OBJECTIVE

We aimed to develop and validate nomogram for the prediction of neonatal mortality. Nomograms are tools which assist the clinical decision making process through early estimation of risks prompting early interventions.

METHODS

A three year retrospective follow up study was conducted at University of Gondar Comprehensive Specialized Hospital and a total of 603 preterm neonates with sepsis were included. Data was collected using KoboCollect and analyzed using STATA version 16 and R version 4.2.1. Lasso regression was used to select the most potent predictors and to minimize the problem of overfitting. Nomogram was developed using multivariable binary logistic regression analysis. Model performance was evaluated using discrimination and calibration. Internal model validation was done using bootstrapping. Net benefit of the nomogram was assessed through decision curve analysis (DCA) to assess the clinical relevance of the model.

RESULT

The nomogram was developed using nine predictors: gestational age, maternal history of premature rupture of membrane, hypoglycemia, respiratory distress syndrome, perinatal asphyxia, necrotizing enterocolitis, total bilirubin, platelet count and kangaroo-mother care. The model had discriminatory power of 96.7% (95% CI: 95.6, 97.9) and P-value of 0.165 in the calibration test before and after internal validation with brier score of 0.07. Based on the net benefit analysis the nomogram was found better than treat all and treat none conditions.

CONCLUSION

The developed nomogram can be used for individualized mortality risk prediction with excellent performance, better net benefit and have been found to be useful in clinical practice with contribution in preterm neonatal mortality reduction by giving better emphasis for those at high risk.

Development and validation of a model to predict mortality risk among extremely preterm infants during the early postnatal period: a multicentre prospective cohort study

BACKGROUND

Recently, with the rapid development of the perinatal medical system and related life-saving techniques, both the short-term and long-term prognoses of extremely preterm infants (EPIs) have improved significantly. In rapidly industrialising countries like China, the survival rates of EPIs have notably increased due to the swift socioeconomic development. However, there is still a reasonably lower positive response towards the treatment of EPIs than we expected, and the current situation of withdrawing care is an urgent task for perinatal medical practitioners.

OBJECTIVE

To develop and validate a model that is practicable for EPIs as soon as possible after birth by regression analysis, to assess the risk of mortality and chance of survival.

METHODS

This multicentre prospective cohort study used datasets from the Sino-Northern Neonatal Network, including 46 neonatal intensive care units (NICUs). Risk factors including maternal and neonatal variables were collected within 1 hour post-childbirth. The training set consisted of data from 41 NICUs located within the Shandong Province of China, while the validation set included data from 5 NICUs outside Shandong Province. A total of 1363 neonates were included in the study.

RESULTS

Gestational age, birth weight, pH and lactic acid in blood gas analysis within the first hour of birth, moderate-to-severe hypothermia on admission and adequate antenatal corticosteroids were influencing factors for EPIs' mortality with important predictive ability. The area under the curve values for internal validation of our prediction model and Clinical Risk Index for Babies-II scores were 0.81 and 0.76, and for external validation, 0.80 and 0.51, respectively. Moreover, the Hosmer-Lemeshow test showed that our model has a constant degree of calibration.

CONCLUSIONS

There was good predictive accuracy for mortality of EPIs based on influencing factors prenatally and within 1 hour after delivery. Predicting the risk of mortality of EPIs as soon as possible after birth can effectively guide parents to be proactive in treating more EPIs with life-saving value.

TRIAL REGISTRATION NUMBER

ChiCTR1900025234.

Association between opioid use during mechanical ventilation in preterm infants and evidence of brain injury: a propensity score-matched cohort study

Background

Preterm infants often require mechanical ventilation (MV), which can be a painful experience. Opioids (such as morphine) are used to provide analgesia, despite conflicting evidence about their impact on the developing brain. We aimed to quantify the use of opioids during MV in infants born at <32 weeks' gestational age and to investigate the association between opioid use and evidence of brain injury.

Methods

In this retrospective propensity score-matched cohort study, we used routinely recorded data from the National Neonatal Research Database to study infants born at 22-31 weeks gestational age who were admitted to neonatal units in England and Wales (between Jan 1, 2012, and Dec 31, 2020) and who were mechanically ventilated on one or more days during their hospital stay. We used propensity score matching to identify pairs of infants (one who received opioids during MV and one who did not) with similar demographic and clinical characteristics. The pre-specified primary outcome was preterm brain injury assessed in all infants who received MV for more than two days and had evidence of preterm brain injury at or before discharge from neonatal care. Adjusted analyses accounted for differences in infants' characteristics, including illness severity and painful/surgical conditions.

Findings

Of 67,206 infants included, 45,193 (67%) were mechanically ventilated for one or more days and 26,201 (58% of 45,193) received an opioid whilst ventilated. Opioids were given for a median of 67% of ventilated days (IQR 43-92%) and the median exposure was 4 days (2-11). The percentage of mechanically ventilated infants who received opioids while ventilated increased from 52% in 2012 to 60% in 2020 (morphine, 51%-56%; fentanyl, 6%-18%). In the propensity score-matched cohort of 3608 pairs who were ventilated for >2 consecutive days, the odds of any preterm brain injury (adjusted odds ratio 1.22, 95% CI 1.10-1.35) were higher in those who received opioids compared with those who did not (received opioids, 990/3608 (27.4%) vs. did not receive opioids, 855/3608 (23.7%). The adjusted odds of these adverse outcomes increased with increasing number of days of opioid exposure.

Interpretation

Use of opioids during mechanical ventilation of preterm infants increased during the study period (2012-2020). Although causation cannot be determined, among those ventilated for >2 consecutive days, these data suggest that opioid use is associated with an increased risk of preterm brain injury and the risk increases with longer durations of exposure.

Funding

University of Nottingham Impact Fund.

Patterns of antibiotic use, pathogens, and prediction of mortality in hospitalized neonates and young infants with sepsis: A global neonatal sepsis observational cohort study (NeoOBS)

BACKGROUND

There is limited data on antibiotic treatment in hospitalized neonates in low- and middle-income countries (LMICs). We aimed to describe patterns of antibiotic use, pathogens, and clinical outcomes, and to develop a severity score predicting mortality in neonatal sepsis to inform future clinical trial design.

METHODS AND FINDINGS

Hospitalized infants <60 days with clinical sepsis were enrolled during 2018 to 2020 by 19 sites in 11 countries (mainly Asia and Africa). Prospective daily observational data was collected on clinical signs, supportive care, antibiotic treatment, microbiology, and 28-day mortality. Two prediction models were developed for (1) 28-day mortality from baseline variables (baseline NeoSep Severity Score); and (2) daily risk of death on IV antibiotics from daily updated assessments (NeoSep Recovery Score). Multivariable Cox regression models included a randomly selected 85% of infants, with 15% for validation. A total of 3,204 infants were enrolled, with median birth weight of 2,500 g (IQR 1,400 to 3,000) and postnatal age of 5 days (IQR 1 to 15). 206 different empiric antibiotic combinations were started in 3,141 infants, which were structured into 5 groups based on the World Health Organization (WHO) AWaRe classification. Approximately 25.9% (n = 814) of infants started WHO first line regimens (Group 1-Access) and 13.8% (n = 432) started WHO second-line cephalosporins (cefotaxime/ceftriaxone) (Group 2-"Low" Watch). The largest group (34.0%, n = 1,068) started a regimen providing partial extended-spectrum beta-lactamase (ESBL)/pseudomonal coverage (piperacillin-tazobactam, ceftazidime, or fluoroquinolone-based) (Group 3-"Medium" Watch), 18.0% (n = 566) started a carbapenem (Group 4-"High" Watch), and 1.8% (n = 57) a Reserve antibiotic (Group 5, largely colistin-based), and 728/2,880 (25.3%) of initial regimens in Groups 1 to 4 were escalated, mainly to carbapenems, usually for clinical deterioration (n = 480; 65.9%). A total of 564/3,195 infants (17.7%) were blood culture pathogen positive, of whom 62.9% (n = 355) had a gram-negative organism, predominantly Klebsiella pneumoniae (n = 132) or Acinetobacter spp. (n = 72). Both were commonly resistant to WHO-recommended regimens and to carbapenems in 43 (32.6%) and 50 (71.4%) of cases, respectively. MRSA accounted for 33 (61.1%) of 54 Staphylococcus aureus isolates. Overall, 350/3,204 infants died (11.3%; 95% CI 10.2% to 12.5%), 17.7% if blood cultures were positive for pathogens (95% CI 14.7% to 21.1%, n = 99/564). A baseline NeoSep Severity Score had a C-index of 0.76 (0.69 to 0.82) in the validation sample, with mortality of 1.6% (3/189; 95% CI: 0.5% to 4.6%), 11.0% (27/245; 7.7% to 15.6%), and 27.3% (12/44; 16.3% to 41.8%) in low (score 0 to 4), medium (5 to 8), and high (9 to 16) risk groups, respectively, with similar performance across subgroups. A related NeoSep Recovery Score had an area under the receiver operating curve for predicting death the next day between 0.8 and 0.9 over the first week. There was significant variation in outcomes between sites and external validation would strengthen score applicability.

CONCLUSION

Antibiotic regimens used in neonatal sepsis commonly diverge from WHO guidelines, and trials of novel empiric regimens are urgently needed in the context of increasing antimicrobial resistance (AMR). The baseline NeoSep Severity Score identifies high mortality risk criteria for trial entry, while the NeoSep Recovery Score can help guide decisions on regimen change. NeoOBS data informed the NeoSep1 antibiotic trial (ISRCTN48721236), which aims to identify novel first- and second-line empiric antibiotic regimens for neonatal sepsis.

TRIAL REGISTRATION

ClinicalTrials.gov, (NCT03721302).

External Validation of the Neonatal Mortality Risk-2000 Score to Predict In-Hospital Mortality in Neonates Weighing 2000 g or Less

The authors aimed to externally validate the Neonatal Mortality Risk-2000 (NMR-2000) score, a simplified tool to predict in-hospital mortality in the setting of a tertiary care hospital. They conducted a single-center prospective cohort study on neonates weighing ≤ 2000 g who were admitted to a neonatal intensive care unit within 6 h of age. The predictors included in the NMR-2000 score were birth weight, SpO₂ at admission, and the highest level of respiratory support during the first 24 h of life. The outcome was in-hospital mortality. Among 243 neonates ≤ 2000 g, there were 94 (38.7%) deaths. The area under the receiver operating characteristic curve value for the NMR score was 0.84 (95% CI 0.79-0.89) in the present sample. The calibration slope was 1, and the intercept was 0. The NMR-2000 score had good discriminating ability and calibration to predict in-hospital mortality.

Neonatal Mortality and Associated Factors in the Neonatal Intensive Care Unit of Gadarif Hospital, Eastern Sudan

Background: Neonatal mortality is a serious public-health issue, especially in Sub-Saharan African countries. There are limited studies on neonatal mortality in Sudan; particularly, there are none on eastern Sudan. Therefore, this study aimed to determine the incidence, causes and associated factors for mortality among neonates admitted to the neonatal intensive care unit (NICU) of Gadarif Hospital, eastern Sudan. Methods: This retrospective study included 543 neonates admitted to the NICU of Gadarif Hospital, eastern Sudan, between January and August 2019. Data were obtained from the hospital record using a questionnaire composed of sociodemographic data, neonatal and maternal information and neonatal outcomes. Logistic regression analyses were performed and the adjusted odds ratio (AOR) and 95% confidence interval (CI) were calculated. Results: Of the 543 neonates, 50.8% were female, 46.4% were low birth weight (LBW), 43.5% were preterm babies and 27% were newborns admitted after caesarean delivery. The neonatal mortality before discharge was 21.9% (119/543) of live-born babies at the hospital. Preterm birth and its complications (48.7%), respiratory distress syndrome (33.6%), birth asphyxia (21.0%) and infection (9.0%) were the most common causes of neonatal mortality. In multivariable logistic regression analysis, preterm birth (AOR 2.10, 95% CI 1.17−3.74), LBW (AOR 2.47, 95% CI 1.38−4.41), low 5 min APGAR score (AOR 2.59, 95% CI 1.35−4.99) and length of hospital stay <3 days (AOR 5.49, 95% CI 3.44−8.77) were associated with neonatal mortality. Conclusion: There is an increased burden of neonatal mortality in the NICU of Gadarif Hospital, eastern Sudan, predominantly among preterm and LBW babies.

External validation of inpatient neonatal mortality prediction models in high-mortality settings

BACKGROUND

Two neonatal mortality prediction models, the Neonatal Essential Treatment Score (NETS) which uses treatments prescribed at admission and the Score for Essential Neonatal Symptoms and Signs (SENSS) which uses basic clinical signs, were derived in high-mortality, low-resource settings to utilise data more likely to be available in these settings. In this study, we evaluate the predictive accuracy of two neonatal prediction models for all-cause in-hospital mortality.

METHODS

We used retrospectively collected routine clinical data recorded by duty clinicians at admission from 16 Kenyan hospitals used to externally validate and update the SENSS and NETS models that were initially developed from the data from the largest Kenyan maternity hospital to predict in-hospital mortality. Model performance was evaluated by assessing discrimination and calibration. Discrimination, the ability of the model to differentiate between those with and without the outcome, was measured using the c-statistic. Calibration, the agreement between predictions from the model and what was observed, was measured using the calibration intercept and slope (with values of 0 and 1 denoting perfect calibration).

RESULTS

At initial external validation, the estimated mortality risks from the original SENSS and NETS models were markedly overestimated with calibration intercepts of - 0.703 (95% CI - 0.738 to - 0.669) and - 1.109 (95% CI - 1.148 to - 1.069) and too extreme with calibration slopes of 0.565 (95% CI 0.552 to 0.577) and 0.466 (95% CI 0.451 to 0.480), respectively. After model updating, the calibration of the model improved. The updated SENSS and NETS models had calibration intercepts of 0.311 (95% CI 0.282 to 0.350) and 0.032 (95% CI - 0.002 to 0.066) and calibration slopes of 1.029 (95% CI 1.006 to 1.051) and 0.799 (95% CI 0.774 to 0.823), respectively, while showing good discrimination with c-statistics of 0.834 (95% CI 0.829 to 0.839) and 0.775 (95% CI 0.768 to 0.782), respectively. The overall calibration performance of the updated SENSS and NETS models was better than any existing neonatal in-hospital mortality prediction models externally validated for settings comparable to Kenya.

CONCLUSION

Few prediction models undergo rigorous external validation. We show how external validation using data from multiple locations enables model updating and improving their performance and potential value. The improved models indicate it is possible to predict in-hospital mortality using either treatments or signs and symptoms derived from routine neonatal data from low-resource hospital settings also making possible their use for case-mix adjustment when contrasting similar hospital settings.

Validation of the STARZ neonatal acute kidney injury risk stratification score

BACKGROUND

Neonatal acute kidney injury (AKI) is common in neonatal intensive care units (NICU) and leads to worse outcomes. Stratifying neonates into an "at risk" category allows health care providers to objectively recognize opportunities for improvements in quality of care.

METHODS

The "Neonatal AKI Risk Prediction Scoring" was devised as the "STARZ [Sethi, Tibrewal, Agrawal, Raina, waZir]" Score. The STARZ score was derived from our prior multicentre study analysing risk factors for AKI in neonates admitted to the NICU. This tool includes 10 variables with a total score ranging from 0 to 100 and a cut-off score of 31.5. In the present study, the scoring model has been validated in our multicentre cohort of 744 neonates.

RESULTS

In the validation cohort, this scoring model had sensitivity of 82.1%, specificity 91.7%, positive predictive value 81.2%, negative predictive value 92.2% and accuracy 88.8%. Based on the STARZ cut-off score of ≥ 31.5, an area under the receiver operating characteristic (ROC) curve was observed to be 0.932 (95% CI, 0.910-0.954; p < 0.001) signifying that the discriminative power was high. In the validation cohort, the probability of AKI was less than 20% for scores up to 32, 20-40% for scores between 33 and 36, 40-60% for scores between 37 and 43, 60-80% for scores between 44 and 49, and ≥ 80% for scores ≥ 50.

CONCLUSIONS

To promote the survival of susceptible neonates, early detection and prompt interventional measures based on highly evidenced research is vital. The risk of AKI in admitted neonates can be quantitatively determined by the rapid STARZ scoring system. A higher resolution version of the Graphical abstract is available as Supplementary information.

Neonatal acute kidney injury risk stratification score: STARZ study

BACKGROUND

Neonates admitted in the neonatal intensive care unit are vulnerable to acute kidney injury leading to worse outcomes. It is important to identify "at-risk" neonates for early preventive measures.

METHODS

The study was a multicenter, national, prospective cohort study done in 11 centers in India. A multivariable logistic regression technique with step-wise backward elimination method was used, and a "Risk Prediction Scoring" was devised [the STARZ score].

RESULTS

The neonates with admission in the NICU within <25.5 h of birth, requirement of positive pressure ventilation in the delivery room, <28 weeks gestational age, sepsis, significant cardiac disease, urine output <1.32 ml/kg/h or serum creatinine ≥0.98 mg/dl during the first 12 h post admission, use of nephrotoxic drugs, use of furosemide, or use of inotrope had a significantly higher risk of AKI at 7 days post admission in the multivariate logistic regression model. This scoring model had a sensitivity of 92.8%, specificity of 87.4% positive predictive value of 80.5%, negative predictive value of 95.6%, and accuracy of 89.4%.

CONCLUSIONS

The STARZ neonatal score serves to rapidly and quantitatively determine the risk of AKI in neonates admitted to the neonatal intensive care unit.

IMPACT

The STARZ neonatal score serves to rapidly and quantitatively determine the risk of AKI in neonates admitted to the neonatal intensive care unit. These neonates with a higher risk stratification score need intense monitoring and daily kidney function assessment. With this intensification of research in the field of AKI risk stratification prediction, there is hope that we will be able to decrease morbidity and mortality associated with AKI in this population.

Risk Prediction for Stillbirth and Neonatal Mortality in Low-resource Settings

High stillbirth and neonatal mortality are major public health problems, particularly in low-resource settings in low- and middle-income countries (LMIC). Despite sustained efforts by national and international organizations over the last several decades, quality intrapartum and neonatal care is not universally available, especially in these low-resource settings. A few studies identify risk factors for adverse perinatal outcomes in low-resource settings in LMICs. This review highlights the evidence of risk prediction for stillbirth and neonatal death. Evidence using advanced machine-learning statistical models built on data from low-resource settings in LMICs suggests that the predictive accuracy for intrapartum stillbirth and neonatal mortality using prenatal and pre-delivery data is low. Models with delivery and post-delivery data have good predictive accuracy of the risk for neonatal mortality. Birth weight is the most important predictor of neonatal mortality. Further validation and testing of the models in other low-resource settings and subsequent development and testing of possible interventions could advance the field.

Observational cohort study of changing trends in non-invasive ventilation in very preterm infants and associations with clinical outcomes

OBJECTIVE

To determine the change in non-invasive ventilation (NIV) use over time in infants born at <32 weeks' gestation and the associated clinical outcomes.

STUDY DESIGN

Retrospective cohort study using routinely recorded data from the National Neonatal Research Database of infants born at <32 weeks admitted to neonatal units in England and Wales from 2010 to 2017.

RESULTS

In 56 537 infants, NIV use increased significantly between 2010 and 2017 (continuous positive airway pressure (CPAP) from 68.5% to 80.2% in 2017 and high flow nasal cannula (HFNC) from 14% to 68%, respectively) (p<0.001)). Use of NIV as the initial mode of respiratory support also increased (CPAP, 21.5%-28.0%; HFNC, 1%-7% (p<0.001)).HFNC was used earlier, and for longer, in those who received CPAP or mechanical ventilation. HFNC use was associated with decreased odds of death before discharge (adjusted OR (aOR) 0.19, 95% CI 0.17 to 0.22). Infants receiving CPAP but no HFNC died at an earlier median chronological age: CPAP group, 22 (IQR 10-39) days; HFNC group 40 (20-76) days (p<0.001). Among survivors, HFNC use was associated with increased odds of bronchopulmonary dysplasia (BPD) (aOR 2.98, 95% CI 2.81 to 3.15) and other adverse outcomes.

CONCLUSIONS

NIV use is increasing, particularly as initial respiratory support. HFNC use has increased significantly with a sevenfold increase soon after birth which was associated with higher rates of BPD. As more infants survive with BPD, we need robust clinical evidence, to improve outcomes with the use of NIV as initial and ongoing respiratory support.

Data Science Trends Relevant to Nursing Practice: A Rapid Review of the 2020 Literature

BACKGROUND

The term "data science" encompasses several methods, many of which are considered cutting edge and are being used to influence care processes across the world. Nursing is an applied science and a key discipline in health care systems in both clinical and administrative areas, making the profession increasingly influenced by the latest advances in data science. The greater informatics community should be aware of current trends regarding the intersection of nursing and data science, as developments in nursing practice have cross-professional implications.

OBJECTIVES

This study aimed to summarize the latest (calendar year 2020) research and applications of nursing-relevant patient outcomes and clinical processes in the data science literature.

METHODS

We conducted a rapid review of the literature to identify relevant research published during the year 2020. We explored the following 16 topics: (1) artificial intelligence/machine learning credibility and acceptance, (2) burnout, (3) complex care (outpatient), (4) emergency department visits, (5) falls, (6) health care-acquired infections, (7) health care utilization and costs, (8) hospitalization, (9) in-hospital mortality, (10) length of stay, (11) pain, (12) patient safety, (13) pressure injuries, (14) readmissions, (15) staffing, and (16) unit culture.

RESULTS

Of 16,589 articles, 244 were included in the review. All topics were represented by literature published in 2020, ranging from 1 article to 59 articles. Numerous contemporary data science methods were represented in the literature including the use of machine learning, neural networks, and natural language processing.

CONCLUSION

This review provides an overview of the data science trends that were relevant to nursing practice in 2020. Examinations of such literature are important to monitor the status of data science's influence in nursing practice.

Machine Learning Algorithms to Predict Mortality of Neonates on Mechanical Intubation for Respiratory Failure

BACKGROUND

Early identification of critically ill neonates with poor outcomes can optimize therapeutic strategies. We aimed to examine whether machine learning (ML) methods can improve mortality prediction for neonatal intensive care unit (NICU) patients on intubation for respiratory failure.

METHODS

A total of 1734 neonates with respiratory failure were randomly divided into training (70%, n = 1214) and test (30%, n = 520) sets. The primary outcome was the probability of NICU mortality. The areas under the receiver operating characteristic curves (AUCs) of several ML algorithms were compared with those of the conventional neonatal illness severity scoring systems including the NTISS and SNAPPE-II.

RESULTS

For NICU mortality, the random forest (RF) model showed the highest AUC (0.939 (0.921-0.958)) for the prediction of neonates with respiratory failure, and the bagged classification and regression tree model demonstrated the next best results (0.915 (0.891-0.939)). The AUCs of both models were significantly better than the traditional NTISS (0.836 (0.800-0.871)) and SNAPPE-II scores (0.805 (0.766-0.843)). The superior performances were confirmed by higher accuracy and F1 score and better calibration, and the superior and net benefit was confirmed by decision curve analysis. In addition, Shapley additive explanation (SHAP) values were utilized to explain the RF prediction model.

CONCLUSIONS

Machine learning algorithms increase the accuracy and predictive ability for mortality of neonates with respiratory failure compared with conventional neonatal illness severity scores. The RF model is suitable for clinical use in the NICU, and clinicians can gain insights and have better communication with families in advance.

Impact of early kangaroo mother care versus standard care on survival of mild-moderately unstable neonates <2000 grams: A randomised controlled trial

BACKGROUND

Understanding the effect of early kangaroo mother care on survival of mild-moderately unstable neonates <2000 g is a high-priority evidence gap for small and sick newborn care.

METHODS

This non-blinded pragmatic randomised clinical trial was conducted at the only teaching hospital in The Gambia. Eligibility criteria included weight <2000g and age 1-24 h with exclusion if stable or severely unstable. Neonates were randomly assigned to receive either standard care, including KMC once stable at >24 h after admission (control) versus KMC initiated <24 h after admission (intervention). Randomisation was stratified by weight with twins in the same arm. The primary outcome was all-cause mortality at 28 postnatal days, assessed by intention to treat analysis. Secondary outcomes included: time to death; hypothermia and stability at 24 h; breastfeeding at discharge; infections; weight gain at 28d and admission duration. The trial was prospectively registered at www.clinicaltrials.gov (NCT03555981).

FINDINGS

Recruitment occurred from 23rd May 2018 to 19th March 2020. Among 1,107 neonates screened for participation 279 were randomly assigned, 139 (42% male [n = 59]) to standard care and 138 (43% male [n = 59]) to the intervention with two participants lost to follow up and no withdrawals. The proportion dying within 28d was 24% (34/139, control) vs. 21% (29/138, intervention) (risk ratio 0·84, 95% CI 0·55 - 1·29, p = 0·423). There were no between-arm differences for secondary outcomes or serious adverse events (28/139 (20%) for control and 30/139 (22%) for intervention, none related). One-third of intervention neonates reverted to standard care for clinical reasons.

INTERPRETATION

The trial had low power due to halving of baseline neonatal mortality, highlighting the importance of implementing existing small and sick newborn care interventions. Further mortality effect and safety data are needed from varying low and middle-income neonatal unit contexts before changing global guidelines.

Development and Validation of a Clinical Prognostic Risk Score to Predict Early Neonatal Mortality, Ethiopia: A Receiver Operating Characteristic Curve Analysis

Background

Early neonatal death is the death of a live-born baby within the first seven days of life, which is 73% of all postnatal deaths in the globe. This study aimed to develop and validate a prognostic clinical risk tool for the prediction of early neonatal death.

Methods

A prospective follow-up study was conducted among 393 neonates at Debre Tabor Referral hospital, Northwest Ethiopia. Multivariable logistic regression model was employed to identify potential prognostic determinants for early neonatal mortality. Area under receiver operating characteristics curve (AUROC) was used to check the model discrimination probability using ‘pROC’ R-package. Model calibration plot was checked using ‘givitiR’ R-package. Finally, a risk score prediction tool was developed for ease of applicability. Decision curve analysis was done for cost-benefit analysis and to check the clinical impact of the model.

Results

Overall, 15.27% (95% CI: 12.03–19.18) of neonates had the event of death during the follow-up period. Maternal undernutrition, antenatal follow-up less than four times, birth asphyxia, low birth weight, and not exclusive breastfeeding were the prognostic predictors of early neonatal mortality. The AUROC for the reduced model was 88.7% (95% CI: 83.8–93.6%), which had good discriminative probability. The AUROC of the simplified risk score algorithm was 87.8% (95% CI, 82.7–92.9%). The sensitivity and specificity of the risk score tool was 70% and 89%, respectively. The true prediction accuracy of the risk score tool to predict early neonatal mortality was 86%, and the false prediction probability was 13%.

Conclusion

We developed an early neonatal death prediction tool using easily available maternal and neonatal characteristics for resource-limited settings. This risk prediction using risk score is an easily applicable tool to identify neonates at a higher risk of having early neonatal mortality. This risk score tool would offer an opportunity to reduce early neonatal mortality, thus improving the overall early neonatal death in a resource-limited setting.

Predicting mortality risk for preterm infants using deep learning models with time-series vital sign data

Mortality remains an exceptional burden of extremely preterm birth. Current clinical mortality prediction scores are calculated using a few static variable measurements, such as gestational age, birth weight, temperature, and blood pressure at admission. While these models do provide some insight, numerical and time-series vital sign data are also available for preterm babies admitted to the NICU and may provide greater insight into outcomes. Computational models that predict the mortality risk of preterm birth in the NICU by integrating vital sign data and static clinical variables in real time may be clinically helpful and potentially superior to static prediction models. However, there is a lack of established computational models for this specific task. In this study, we developed a novel deep learning model, DeepPBSMonitor (Deep Preterm Birth Survival Risk Monitor), to predict the mortality risk of preterm infants during initial NICU hospitalization. The proposed deep learning model can effectively integrate time-series vital sign data and fixed variables while resolving the influence of noise and imbalanced data. The proposed model was evaluated and compared with other approaches using data from 285 infants. Results showed that the DeepPBSMonitor model outperforms other approaches, with an accuracy, recall, and AUC score of 0.888, 0.780, and 0.897, respectively. In conclusion, the proposed model has demonstrated efficacy in predicting the real-time mortality risk of preterm infants in initial NICU hospitalization.

Hybridized neural networks for non-invasive and continuous mortality risk assessment in neonates

Premature birth is the primary risk factor in neonatal deaths, with the majority of extremely premature babies cared for in neonatal intensive care units (NICUs). Mortality risk prediction in this setting can greatly improve patient outcomes and resource utilization. However, existing schemes often require laborious medical testing and calculation, and are typically only calculated once at admission. In this work, we propose a shallow hybrid neural network for the prediction of mortality risk in 3-day, 7-day, and 14-day risk windows using only birthweight, gestational age, sex, and heart rate (HR) and respiratory rate (RR) information from a 12-h window. As such, this scheme is capable of continuously updating mortality risk assessment, enabling analysis of health trends and responses to treatment. The highest performing scheme was the network that considered mortality risk within 3 days, with this scheme outperforming state-of-the-art works in the literature and achieving an area under the receiver-operator curve (AUROC) of 0.9336 with standard deviation of 0.0337 across 5 folds of cross-validation. As such, we conclude that our proposed scheme could readily be used for continuously-updating mortality risk prediction in NICU environments.

Prediction modelling of inpatient neonatal mortality in high-mortality settings

OBJECTIVE

Prognostic models aid clinical decision making and evaluation of hospital performance. Existing neonatal prognostic models typically use physiological measures that are often not available, such as pulse oximetry values, in routine practice in low-resource settings. We aimed to develop and validate two novel models to predict all cause in-hospital mortality following neonatal unit admission in a low-resource, high-mortality setting.

STUDY DESIGN AND SETTING

We used basic, routine clinical data recorded by duty clinicians at the time of admission to derive (n=5427) and validate (n=1627) two novel models to predict in-hospital mortality. The Neonatal Essential Treatment Score (NETS) included treatments prescribed at the time of admission while the Score for Essential Neonatal Symptoms and Signs (SENSS) used basic clinical signs. Logistic regression was used, and performance was evaluated using discrimination and calibration.

RESULTS

At derivation, c-statistic (discrimination) for NETS was 0.92 (95% CI 0.90 to 0.93) and that for SENSS was 0.91 (95% CI 0.89 to 0.93). At external (temporal) validation, NETS had a c-statistic of 0.89 (95% CI 0.86 to 0.92) and SENSS 0.89 (95% CI 0.84 to 0.93). The calibration intercept for NETS was -0.72 (95% CI -0.96 to -0.49) and that for SENSS was -0.33 (95% CI -0.56 to -0.11).

CONCLUSION

Using routine neonatal data in a low-resource setting, we found that it is possible to predict in-hospital mortality using either treatments or signs and symptoms. Further validation of these models may support their use in treatment decisions and for case-mix adjustment to help understand performance variation across hospitals.

Effect of early kangaroo mother care on time to full feeds in preterm infants - A prospective cohort study

BACKGROUND

Kangaroo mother care (KMC) is known to reduce neonatal mortality and morbidity. In preterm neonates, KMC is usually initiated only after stabilization.

AIMS

We aimed to assess if early initiation of KMC starting within the first week of life is safe, and reduces the time to full feeds (TFF) in preterm neonates.

STUDY DESIGN

Prospective cohort study.

SUBJECTS

Preterm neonates (Gestation ≤ 34 weeks, Birth weight ≤ 1250 g). This was studied in two epochs, (epoch 1) which was before early KMC vs. epoch 2 which was after implementation of early KMC even if they needed respiratory support, with umbilical/central lines in situ.

OUTCOME

The primary outcome of the study was time to establish full feeds (TFF) of 150 ml/kg/day.

RESULTS

The neonatal demographic characteristics were comparable between epoch 1 and epoch 2 except for lower gestational age, higher surfactant, and any respiratory support in epoch 2. On univariate analysis, early KMC significantly reduced TFF (12.5 vs. 9 days, P < 0.001). Feed intolerance, duration of parenteral nutrition were significantly reduced, and discharge weight Z score improved significantly in epoch 2. On multivariate regression analysis early KMC, exclusive mother's own milk feeding and blood culture-positive late-onset sepsis were important predictors of TFF. Early KMC was safe and well-tolerated.

CONCLUSION

Early KMC was safe and associated with reduced TFF and other nutritional benefits in moderately ill preterm neonates.

Burden of disease and risk factors for mortality amongst hospitalized newborns in Nigeria and Kenya

OBJECTIVE

To describe the patient population, priority diseases and outcomes in newborns admitted <48 hours old to neonatal units in both Kenya and Nigeria.

STUDY DESIGN

In a network of seven secondary and tertiary level neonatal units in Nigeria and Kenya, we captured anonymised data on all admissions <48 hours of age over a 6-month period.

RESULTS

2280 newborns were admitted. Mean birthweight was 2.3 kg (SD 0.9); 57.0% (1214/2128) infants were low birthweight (LBW; <2.5kg) and 22.6% (480/2128) were very LBW (VLBW; <1.5 kg). Median gestation was 36 weeks (interquartile range 32, 39) and 21.6% (483/2236) infants were very preterm (gestation <32 weeks). The most common morbidities were jaundice (987/2262, 43.6%), suspected sepsis (955/2280, 41.9%), respiratory conditions (817/2280, 35.8%) and birth asphyxia (547/2280, 24.0%). 18.7% (423/2262) newborns died; mortality was very high amongst VLBW (222/472, 47%) and very preterm infants (197/483, 40.8%). Factors independently associated with mortality were gestation <28 weeks (adjusted odds ratio 11.58; 95% confidence interval 4.73-28.39), VLBW (6.92; 4.06-11.79), congenital anomaly (4.93; 2.42-10.05), abdominal condition (2.86; 1.40-5.83), birth asphyxia (2.44; 1.52-3.92), respiratory condition (1.46; 1.08-2.28) and maternal antibiotics within 24 hours before or after birth (1.91; 1.28-2.85). Mortality was reduced if mothers received a partial (0.51; 0.28-0.93) or full treatment course (0.44; 0.21-0.92) of dexamethasone before preterm delivery.

CONCLUSION

Greater efforts are needed to address the very high burden of illnesses and mortality in hospitalized newborns in sub-Saharan Africa. Interventions need to address priority issues during pregnancy and delivery as well as in the newborn.

Non-linear association between admission temperature and neonatal mortality in a low-resource setting

Both neonatal hypothermia and hyperthermia represent important risk factors for neonatal mortality, but information on mortality risk across a full range of neonatal temperatures is lacking in low-resource settings. We evaluated the association between neonatal mortality and a full range of admission temperatures in a low-resource setting. This retrospective observational study was conducted at Beira Central Hospital, Mozambique. The relationship between admission temperature and mortality was evaluated using multivariable analyses with temperature modeled as non-linear term. Among 2098 neonates admitted to the Special Care Unit between January–December 2017, admission temperature was available in 1344 neonates (64%) who were included in the analysis. A non-linear association between mortality rate and temperature was identified. Mortality rate decreased from 84% at 32 °C to 64% at 34.6 °C (− 8% per °C), to 41% at 36 °C (− 16% per °C), to 26% to 36.6 °C (− 25% per °C) and to 22% at 38.3 °C (− 2% per °C), then increased to 40% at 41 °C (+ 7% per °C). Mortality rate was estimated to be at minimum at admission temperature of 37.5 °C. In conclusions, the non-linear relationship highlighted different mortality risks across a full range of neonatal temperatures in a low-resource setting. Admission temperature was not recorded in one third of neonates.

Predictive Modeling for Perinatal Mortality in Resource-Limited Settings

IMPORTANCE

The overwhelming majority of fetal and neonatal deaths occur in low- and middle-income countries. Fetal and neonatal risk assessment tools may be useful to predict the risk of death.

OBJECTIVE

To develop risk prediction models for intrapartum stillbirth and neonatal death.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study used data from the Eunice Kennedy Shriver National Institute of Child Health and Human Development Global Network for Women's and Children's Health Research population-based vital registry, including clinical sites in South Asia (India and Pakistan), Africa (Democratic Republic of Congo, Zambia, and Kenya), and Latin America (Guatemala). A total of 502 648 pregnancies were prospectively enrolled in the registry.

EXPOSURES

Risk factors were added sequentially into the data set in 4 scenarios: (1) prenatal, (2) predelivery, (3) delivery and day 1, and (4) postdelivery through day 2.

MAIN OUTCOMES AND MEASURES

Data sets were randomly divided into 10 groups of 3 analysis data sets including training (60%), test (20%), and validation (20%). Conventional and advanced machine learning modeling techniques were applied to assess predictive abilities using area under the curve (AUC) for intrapartum stillbirth and neonatal mortality.

RESULTS

All prenatal and predelivery models had predictive accuracy for both intrapartum stillbirth and neonatal mortality with AUC values 0.71 or less. Five of 6 models for neonatal mortality based on delivery/day 1 and postdelivery/day 2 had increased predictive accuracy with AUC values greater than 0.80. Birth weight was the most important predictor for neonatal death in both postdelivery scenarios with independent predictive ability with AUC values of 0.78 and 0.76, respectively. The addition of 4 other top predictors increased AUC to 0.83 and 0.87 for the postdelivery scenarios, respectively.

CONCLUSIONS AND RELEVANCE

Models based on prenatal or predelivery data had predictive accuracy for intrapartum stillbirths and neonatal mortality of AUC values 0.71 or less. Models that incorporated delivery data had good predictive accuracy for risk of neonatal mortality. Birth weight was the most important predictor for neonatal mortality.