Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jun 18.
Published in final edited form as: JAMA Pediatr. 2026 Apr 1;180(4):384–393. doi: 10.1001/jamapediatrics.2025.5967

Clinical Signs Associated with Mortality and Sepsis in Young Infants: A Systematic Review and Meta-analysis

Sophie Driker 1, Sitarah Mathias 2,3, Alastair Fung 4,11, Suci Ardini Widyaningsih 1,3, Naomi Schmeck 1,5, Jana Adnan 6,7, Yumin Kim 8, Anum S Hussaini 1,5, Tessa Kehoe 1, Krysten North 1,3, Amber Hoey 9, Yasir Shafiq 10,11, Carrie G Wade 12, Rishi P Mediratta 13, Chris A Rees 14, Anne CC Lee 8,1
PMCID: PMC12865698  NIHMSID: NIHMS2178998  PMID: 41627835

Abstract

Importance:

Early and accurate identification of clinical warning signs in young infants may help avert sepsis morbidity and mortality in resource-limited settings.

Objective:

To systematically review evidence on the association and accuracy of clinical signs to diagnose sepsis or predict mortality in young infants aged 0 to 59 days to inform management in settings with limited laboratory diagnostics.

Data Sources:

Medline, Embase, CINAHL, Global Index Medicus, and Cochrane CENTRAL Register were searched from inception through May 2023, with updated searches on September 5, 2024. An umbrella search of systematic reviews was conducted in January 2024.

Study Selection:

Included studies reported data on 24 infant clinical signs informed by current World Health Organization (WHO) Integrated Management of Childhood Illness (IMCI) and hospital-based algorithms for the care of sick young infants reporting odds ratios (OR), risk ratios, or sensitivity and specificity.

Data Extraction and Synthesis:

Data were extracted independently by 2 reviewers. Quality assessment used the Newcastle-Ottawa, Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2), and Quality Assessment of Prognostic Accuracy Studies (QUAPAS) Scales. OR data were pooled using random-effects models. Data analysis was performed from July to September 2025.

Main Outcomes and Measures:

OR of all-cause mortality, culture-confirmed sepsis, or clinical sepsis (with access to laboratory investigations).

Results:

Of 7,641 studies, 52 studies with 140,885 participants were included. A total of 16 clinical signs were significantly associated with mortality, 11 with culture-confirmed sepsis, and 13 with clinical sepsis. For mortality, the five strongest associations were weak, abnormal or absent cry (OR, 20.48; 95% CI, 6.59–63.67); not able to feed at all (OR, 18.32; 95% CI, 6.00–55.97); not feeding well (OR, 13.39; 95% CI, 6.97–25.72); drowsiness or unconsciousness (OR, 12.46; 95% CI, 6.06–25.62); and prolonged capillary refill (OR, 12.06; 95% CI, 2.77–52.53). The top 5 signs associated with culture-confirmed sepsis were not feeding well (OR, 4.52; 95% CI, 1.10–18.59), prolonged capillary refill (OR, 3.59; 95% CI, 2.05–6.28), lethargy (OR, 3.44; 95% CI, 1.89–6.26), drowsiness or unconsciousness (OR, 3.07; 95% CI, 2.01–4.68), and feeding intolerance (OR, 2.95; 95% CI, 1.67–5.21).

Conclusions and Relevance:

All current WHO IMCI clinical signs were significantly associated with mortality or culture-confirmed sepsis. Several signs not in IMCI were identified that may improve identification of life-threatening illness in young infants in resource-limited settings where clinical sign algorithms are the primary diagnostic tool.

BACKGROUND/INTRODUCTION

Morbidity and mortality in young infants (<60 days) remain an urgent global health challenge with a disproportionate burden in low- and middle-income countries (LMICs).1-3 Approximately 2.3 million neonates die each year, with greater than half a million deaths attributable to serious bacterial infections.2,4 Serious bacterial infections, such as sepsis, meningitis, and pneumonia can cause long-term disability among surviving infants, contributing to approximately 3% of all disability-adjusted life years worldwide.5 The World Health Organization (WHO) has identified a critical need for improved approaches to accurately identify infants requiring treatment of serious bacterial infections.2

In LMICs where resources and advanced diagnostic tools are scarce, the WHO’s Integrated Management of Childhood Illness (IMCI) was developed in 1995 to improve the diagnosis and management of major illnesses in settings with limited healthcare infrastructure.6 IMCI has been adopted in over 100 countries worldwide7 and employs a clinical sign “checklist” for frontline health workers to enable timely identification of possible serious bacterial infections (pSBI) and need for antibiotic treatment. The Young Infant Clinical Signs study (YIS-2),8 informed the development of the seven-sign IMCI pSBI algorithm and current guidelines for management of sick infants presenting to health facilities, published in 2019.9 Another algorithm, the Pocket Book of Hospital Care for Children, introduced five additional signs for hospital-based management.10 Despite limitations of using only infant clinical signs to detect serious bacterial infections in young infants,2,11 they often remain the first indication of severe illness in community or primary facilities lacking higher-level diagnostic capacity.12,13 Given IMCI’s widespread use, assessing the performance of different clinical signs is critical to ensure optimal diagnostic accuracy and timely management of illness to prevent mortality in young infants.

We conducted this systematic review and meta-analysis to synthesize evidence on the accuracy of infant clinical signs to predict mortality and sepsis, examining the following PIRD (Population, Index test, Reference Standard, Diagnosis of interest) question: Among young infants aged 0-59 days at presentation, in any setting, what is the association of clinical infant signs with sepsis or mortality from any cause by 60 days of life?

METHODS

Study Design

We conducted a systematic review and meta-analysis based on a registered PROSPERO protocol (CRD42023431387)14 according to Cochrane15 and PRISMA-2020 reporting guidelines (eTable 1).16 As no human subjects data were collected, we did not obtain ethical approval.

Definitions

Exposures:

We defined a clinical sign as a postnatal infant sign or symptom ascertained by physical examination or history. We included 24 clinical signs from the WHO IMCI algorithm for clinical severe infection,9 WHO IMCI algorithm for critical illness,9 WHO Pocket Book of Hospital Care for Children,10 and the YIS-2 study8 which informed the current 7-sign IMCI.8 Two signs, vomiting and feeding intolerance, were selected by the authors based on clinical relevance (Table 1). As exact sign definitions varied by study, we categorized similar signs to facilitate analyses. Signs were assumed to be directly observed by study staff unless otherwise stated by the authors.

Table 1.

Signs included in this review and current World Health Organization (WHO) clinical algorithms from which signs were identified*

 WHO Clinical Signs
 of Clinical Severe
 Infection
 (6 signs, 2019)		  WHO Clinical Signs
 of Critical Illness

 (3 signs, 2019)		  WHO Pocket Book
 of Hospital Care for
 Children
 (12 signs, 2013)		  WHO Clinical Signs of
 Possible Serious
 Bacterial Infection
 “Current 7-sign IMCI”
 (7 signs, 2019)		  Additional clinical
 signs from YIS-2
 study
 (9 signs, 2008)
•not feeding well • not able to feed at all •not feeding well •not feeding well or not able to feed at all •restless and irritable
•movement only when stimulated • no movement at all •movement only when stimulated or no movement at all •movement only when stimulated or no movement at all • inconsolable
•high body temperature (≥38°C) • convulsions •high body temperature (>38°) •high body temperature (≥38°C) •weak abnormal or absent cry
•low body temperature (<35.5°C) •low body temperature (<35.5°C) •low body temperature (<35.5°C) •prolonged capillary refill or poor peripheral perfusion
•severe chest indrawing •severe chest indrawing •severe chest indrawing •reduced skin turgor
•fast breathing (≥ 60 breaths per minute) in infants aged 0-6 days •fast breathing (≥ 60 breaths per minute) •fast breathing (≥ 60 breaths per minute) in infants aged 0-6 days •nasal flaring
•convulsions •convulsions •bulging fontanelle
•drowsy or unconsciousness •vomiting**
•grunting •feeding intolerance**
•central cyanosis
•severe jaundice
•severe abdominal distension
Open in a new tab

IMCI= Integrated Management of Childhood Illness, WHO= World Health Organization

*

All algorithms specify that young infants with one or more signs should be referred to hospital urgently for treatment

**

Vomiting and feeding intolerance were identified not from YIS-2 study but from authors’ clinical judgment

Clinical Outcomes:

Mortality and sepsis (culture-confirmed or clinical) were primary outcomes. Mortality was defined as death from any cause within 60 days of life or until hospital discharge. As some studies used the term “culture-confirmed” while actually including both blood culture confirmed and non blood culture confirmed cases, we required at least 50% of cases having positive blood culture to be classified as culture-confirmed sepsis. Clinical sepsis was defined as clinician judgment of suspected or confirmed serious bacterial infection supported by laboratory investigations or imaging (e.g., positive urine, cerebrospinal fluid, hematological markers, x-ray).

Prognostic/Diagnostic Outcome Measures:

The primary measure of association was odds ratios (OR). Diagnostic accuracy measures of sensitivity and specificity were secondary outcomes. Studies reporting risk ratio (RR) were also included.

Search Strategy

Our search strategy included terms for neonates/infants, sepsis, specific clinical symptoms/signs, and diagnostic accuracy measures (eTable 2a). A medical research librarian (CW) searched MEDLINE, Embase, CINAHL, Global Index Medicus, and Cochrane CENTRAL Registry of Trials from inception to May 2023, and updated the searches in September 5, 2024. An umbrella search of systematic reviews was conducted in January 2024 to identify additional records (eTable 2b). We also hand-searched bibliographies of relevant reviews.

Study Selection

Records were imported into Covidence and deduplicated.17 Two reviewers independently screened studies by title/abstract, then full text. A third reviewer resolved disagreements.

Inclusion and Exclusion Criteria

We included studies that: (1) analyzed infants aged 0 to 59 days, or with a mean or median age less than 60 days; (2) reported the association of a postnatal clinical sign with an available reference standard; and (3) reported a pre-specified prognostic/diagnostic outcome measure within main paper results. We excluded studies: (1) limited to specialized populations (e.g., all infants underwent surgery); (2) of prenatal or intrapartum signs risk factors, laboratory tests, biomarkers, or Apgar scores alone; (3) with signs not routinely ascertainable in primary or secondary care;18 (4) using physician clinical judgment of illness alone, without laboratory or imaging, as the reference standard; (5) reviews, conference proceedings, protocols, case reports, commentaries; and (6) with subpopulations of studies included in this review.

Data Extraction and Management

Two reviewers independently extracted data into an Excel form. For each clinical sign, we extracted all reported outcome measures when sufficient data were available.

Quality Assessment

We used the Newcastle Ottawa Scale19 for observational studies reporting OR/RR, the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2)20 for diagnostic accuracy studies on sepsis, and the Quality Assessment of Prognostic Accuracy Studies (QUAPAS)21 for diagnostic accuracy studies on mortality. Two reviewers assessed studies independently; a third resolved disagreements.

The Newcastle Ottawa Scale was adapted for our review, with detailed decision rules provided in eTables 6-7. Studies were graded as good, fair, or poor according to Agency for Healthcare Research and Quality standards.22 QUADAS-2 and QUAPAS assessed quality across domains (participant selection, index test, reference standard, flow and timing, and analysis for QUAPAS), rating studies as high, unclear, or low quality. Both tools, detailed in eTables 8-9, also assessed applicability. We did not assess publication bias due to insufficient data.15

Statistical Analysis

We conducted pooled analyses using Stata SE 18 (StataCorp, College Station, TX).23 To pool diagnostic accuracy measures, we calculated 2-by-2 tables (true positives, false positives, true negatives, and false negatives) from available data. We excluded data points from meta-analysis if calculated values were greater than 30% different from paper’s reported OR or greater than 5% different from sensitivity and specificity, or could not be reconciled with the original paper.

We used crude ORs for pooling, including adjusted ORs only if crude results were unavailable. We log-transformed values and generated DerSimonian Laird random-effects models.24,25 To pool diagnostic accuracy, we generated hierarchical summary receiver operating characteristic curves accounting for the correlation between sensitivity and specificity where there were at least 4 studies, following Cochrane methods.26,27 When there were fewer than 4 studies or convergence failed, we used random-effects bivariate models.28 We reported stratified associations among directly observed versus self-reported signs. We also conducted sensitivity analyses restricted to studies in LMICs.

RESULTS

Study Characteristics

Of 7,641 studies identified and 28 from citation searching, 52 met the inclusion criteria (Figure 1).8,29-79 Reasons for exclusion at the full-text stage are provided in eTable 3. Characteristics of studies excluded for reporting only hazard ratios are in eTable 4. Characteristics of included studies are shown in eTable 5.

Figure 1. PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and other sources.

Figure 1.

Open in a new tab

PRISMA figure template from: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021 ;372:n71 .doi: 10.1136/bmj.n71.

Studies were published between 1991 and 2024, and one 2025 preprint. 43 were prospective or retrospective observational studies, 5 were case-control, and 4 were randomized-controlled trials. Sample sizes ranged from 47 to 6,924 (total n=140,885). Studies were conducted in 30 countries, including 9 high-income and 21 LMICs. Six studies were in community settings, 13 in outpatient hospital or primary facilities, and 33 in inpatient hospitals or specialized units such as neonatal intensive care units (NICUs).

Methodological Quality of Included Studies

Newcastle Ottawa Scale assessments are shown in eFigures 1-2. Among 35 studies reporting OR/RR, 28 (80%) were “good”, 6 (17%) were “fair”, and 1 (3%) was “poor” quality. QUADAS-2 and QUAPAS assessments are reported in eFigures 3-4. Among 14 studies assessed using QUADAS-2, 12 were assigned “not serious” and two were “serious”. Among three studies assessed with QUAPAS, two were “not serious” and one was “serious”.

Results of pooled analyses

Of 24 directly-observed signs with poolable OR data; 16 signs were significantly associated with mortality, 11 with culture-confirmed sepsis, and 13 with clinical sepsis. Study-level results are in eTable 10. Sensitivity and specificity were pooled for 10 signs for mortality, 12 for culture-confirmed sepsis, and 16 for clinical sepsis (eTable 11). Pooled results of the sensitivity analysis of studies in LMICs were largely comparable to the main analysis (eFigures 5-7).

Pooled results for mortality

16 signs were significantly associated with mortality, and the five clinical signs most strongly associated with mortality had ORs greater than 12.0: weak abnormal or absent cry (OR, 20.48; 95% CI, 6.59–63.67; 3 studies); not able to feed at all (OR, 18.32; 95% CI, 6.00–55.97; 2 studies); not feeding well (OR, 13.39; 95% CI, 6.97–25.72; 4 studies); drowsy or unconsciousness (OR, 12.46; 95% CI, 6.06–25.62; 5 studies); and prolonged capillary refill (OR, 12.06; 95% CI, 2.77–52.53; 3 studies) (Figure 2). The most sensitive signs were not feeding well (74.4%; 95% CI, 28.0–95.6%), low body temperature (66.1%; 95% CI, 40.3–85.0%) and weak abnormal or absent cry (47.2%; 95% CI, 38.5–56.0%). The most specific signs were severe jaundice (99.1%; 95% CI, 98.8–99.3%), drowsy or unconsiousness (98.8%; 95% CI, 97.5–99.4%) and central cyanosis (97.8%; 95% CI, 92.9–99.4%) (eTable 11). Among seven signs included in the current IMCI algorithm, six were significantly associated with increased odds of mortality, while convulsions showed a positive but nonsignificant association.

Figure 2. Pooled association of clinical signs with culture-confirmed sepsis.

Figure 2.

Open in a new tab

*Sign in current IMCI-7 algorithm

Pooled results for culture-confirmed sepsis

Eleven clinical signs were significantly associated with culture-confirmed sepsis. The five strongest associations were: not feeding well (OR, 4.52; 95% CI, 1.10–18.59; 3 studies), prolonged capillary refill (OR, 3.59; 95% CI, 2.05–6.28; 7 studies), lethargy (OR, 3.44; 95% CI, 1.89–6.26; 8 studies), drowsy or unconsciousness (OR, 3.07; 95% CI, 2.01–4.68; 2 studies), and feeding intolerance (OR, 2.95; 95% CI, 1.67-5.21, 4 studies) (Figure 3). The most sensitive signs were lethargy (47.8%; 95% CI, 34.5-61.5%), prolonged capillary refill (43.7%; 95% CI, 23.4-66.3%), and not able to feed at all (38.8%; 95% CI, 17.8-64.9%). The most specific were grunting (95.7%; 95% CI, 91.3-97.9%), convulsions (94.6%; 95% CI, 88.4-97.6%), and high body temperature (91.9%; 95% CI, 72.3-98.0%) (eTable 11). Among the current IMCI, six of seven signs were significantly associated with culture-confirmed sepsis, except fast breathing (OR, 1.09; 95% CI, 0.73–1.63; 8 studies).

Figure 3. Pooled association of clinical signs with mortality.

Figure 3.

Open in a new tab

*Sign in current IMCI-7 algorithm

Pooled results for clinical sepsis

Thirteen signs were significantly associated with clinical sepsis (Figure 4), with the five strongest associations being movement only when stimulated (OR, 11.00; 95% CI, 3.40–35.58; 2 studies), central cyanosis (OR, 7.35; 95% CI, 3.19–16.93; 4 studies), bulging fontanelle (OR, 6.79; 95% CI, 3.64–12.67; 2 studies), nasal flaring (OR, 6.08; 95% CI, 2.26–16.36; 2 studies), and not feeding well (OR, 5.11; 95% CI, 2.77–9.43; 5 studies). The top 3 most sensitive signs were not feeding well (56.0%; 95% CI, 44.2-67.3%), high body temperature (49.3%; 95%, 31.8-67.0%), and fast breathing (48.7%; 95 % CI, 31.1-66.6%). The top 3 most specific were central cyanosis (98.2%; 95% CI, 92.7-99.6%), drowsy or unconsciousness (97.6%; 95% CI, 95.7-98.6%), and vomiting (94.8%; 93.3-95.9%) (eTable 11). All 7 IMCI signs were significantly associated with clinical sepsis, except convulsions, which was excluded due to insufficient data.

Figure 4. Pooled association of clinical signs with clinical sepsis.

Figure 4.

Open in a new tab

*Sign in current IMCI-7 algorithm

DISCUSSION

This systematic review comprehensively examines the association of different infant clinical warning signs with mortality and sepsis. All clinical signs from the current WHO IMCI possible SBI algorithm were significantly associated with adverse outcomes of mortality, culture-confirmed sepsis, or clinical sepsis in young infants aged 0 to 59 days. We also identified several predictive signs not in IMCI, including prolonged capillary refill and central cyanosis, suggesting opportunities to improve diagnostic accuracy of WHO IMCI algorithms. A previous review by Opiyo and English assessed clinical signs of severe illness among young infants in LMICs but was limited to a narrative summary of 5 included studies.80 We have previously reported upon the diagnostic accuracy of clinical sign algorithms,81 whereas this systematic review addresses a gap in current literature by meta-analyzing 24 individual clinical signs with mortality, culture-confirmed sepsis, and clinical sepsis to identify opportunities to improve current IMCI recommendations.

All signs included in the current IMCI algorithm demonstrated significant associations with adverse outcomes, supporting their continued inclusion in clinical protocols. Poor feeding was a top predictor of adverse outcomes and had the highest sensitivity for mortality (74.4%) and clinical sepsis (56.0%). Infants require adequate energy for nervous system coordination of suckling, swallowing and breathing: poor feeding may be among the earliest signs of infection and metabolic stress.82 This aligns with findings of a large multi-site study in African countries (AFRINEST), which found the highest mortality rates for infants who were unable to feed at all among various IMCI signs.83 Hypothermia was associated with four-fold increased odds of mortality, consistent with several other studies.83-85 In a forthcoming review, we systematically review the optimal thresholds for fever and hypothermia.86 Notably, although IMCI includes “no movement or movement only when stimulated” as an indicator of altered consciousness, our analysis identified additional neurological signs and descriptions of altered consciousness with high odds: lethargy, drowsy or unconsciousness, and weak abnormal or absent cry. Future studies may assess and compare the accuracy of these different descriptors within the same population to select the optimal description of altered consciousness.

Several IMCI signs were not significantly associated with mortality and/or sepsis in this review. Convulsions were associated with culture-confirmed sepsis but not significantly associated with mortality, possibly due to the low prevalence of convulsions and/or reliance on maternal reports. Fast breathing, or tachypnea, was significantly associated with mortality but not with culture-confirmed bacterial sepsis across 8 studies (n=52,222). Additionally, tachypnea may indicate serious viral respiratory infections, such as respiratory syncytial virus, that can be life-threatening in young infants and require close monitoring and supportive care, even when antibiotic therapy is not indicated. Respiratory rate in young infants may also be influenced by non-infectious conditions (prematurity, respiratory distress syndrome, transient tachypnea of the newborn), thus reducing the diagnostic accuracy of tachypnea alone as a marker of bacterial infection.87,88

Prolonged capillary refill, which is not in current WHO algorithms, was significantly associated with adverse outcomes and emerged as the second-most sensitive individual sign for culture-confirmed sepsis (43.7%). Studies used varying definitions (2 or 3 seconds) and measurement sites (finger, foot, or sternum), potentially reflecting the lack of standardized clinical guidelines for assessment. Despite heterogeneity, prolonged capillary refill demonstrated stronger associations with mortality and sepsis than several IMCI signs, including temperature abnormalities and tachypnea, and showed relatively higher sensitivity for culture-confirmed sepsis (43.7%) and clinical sepsis (40.0%), with specificity of 83.0% and 75.4%, respectively. These findings align with a 2015 meta-analysis that found four-fold increased mortality risk in children with prolonged capillary refill.89 Prolonged capillary refill was observed in more than 10% of infants in pooled studies for mortality and culture-confirmed sepsis, and previous WHO hospital guidelines have recognized prolonged capillary refill as a sign of shock requiring immediate treatment.90 Future studies should consider validation and inclusion of prolonged capillary refill in algorithms, with clear standardization of its definition and measurement method.89 Current practice guidelines endorse the use of a 3-second cutoff and measurement with a digital stopwatch to improve inter-rater reliability.91 Certain factors, including early neonatal age, colder ambient temperatures, and darker skin pigmentation, should be considered in future validation studies of prolonged capillary refill.92

Central cyanosis, a current sign in the WHO Pocket Book of Hospital Care for Children, was significantly associated with mortality and sepsis with higher odds than other respiratory distress signs (tachypnea, severe chest indrawing, nasal flaring, grunting). Although central cyanosis may be a more severe sign and less frequently observed (prevalence around 3%), future algorithms may investigate the comparative accuracy of central cyanosis versus currently-included respiratory signs in IMCI. Importantly, central cyanosis is a subjective finding that may be more difficult to appreciate in deeply pigmented infants; Variability in assessment may be mitigated by developing and validating specific methods, such as assessment of the internal oral mucosa instead of external skin.93

Our analysis has important implications. Global under-five mortality rates have stagnated since 2015, and reducing neonatal and under-5 mortality remains a Sustainable Development Goal.94 Current IMCI checklist algorithms to identify possible serious bacterial infections have moderate sensitivity and specificity;12 We identified several high-performing signs not currently included in WHO IMCI that may improve diagnostic accuracy. However, algorithm revision must balance diagnostic performance with practical implementation in LMIC settings. The 7-sign IMCI checklist was designed to optimize efficiency while maintaining adequate sensitivity, as adding numerous signs increases training complexity, assessment time, and resource requirements for community health workers.95 WHO approaches must prioritize clinical signs that are objective, feasible, and have adequate inter-rater reliability across diverse populations. It remains uncertain whether adding the candidate signs identified in this review would meaningfully improve risk stratification beyond current WHO recommendations. Prospective validation studies are needed to determine whether incorporating these signs into clinical algorithms provides incremental predictive accuracy, can be reliably assessed in practice, and leads to better clinical outcomes. Alternative approaches, such as regression or machine learning models, may also warrant investigation.

Strengths of this study include a relatively large amount of data from LMICs, which bear the disproportionate burden of poor outcomes. Studies were generally good quality despite limited data for some signs. Limitations of this study include heterogeneity of clinical signs and reference standard definitions, as well as considerable variation in terminologies, measurement, and thresholds of signs. We presented only directly observed signs in main results and detail caregiver-reported data separately. Data were observational, and we did not control for potential confounding factors and are unable to determine the influence of other characteristics such as delivery and maternal conditions, gestational age, or birth size.

CONCLUSIONS

Results of this systematic review and meta-analysis show that all 7 clinical signs included in the current WHO IMCI possible SBI algorithm were significantly associated with sepsis and mortality in young infants aged 0 to 59 days. Several signs with the strongest associations, including prolonged capillary refill and central cyanosis, are not currently included in IMCI. Future research may test the validity of incorporating these clinical signs and using non-checklist methods, such as machine learning, to optimize algorithms and effectively manage possible serious bacterial infections in the young infant period.

Supplementary Material

Supplementary material
Supplementary tables

Key Points.

Question:

Which clinical signs are most strongly associated with mortality and sepsis in young infants aged 0–59 days?

Findings:

In this systematic review and meta-analysis of 52 studies including 140,885 participants, all clinical signs in the World Health Organization (WHO)’s current Integrated Management of Childhood Illness (IMCI) algorithm were significantly associated with mortality or culture-confirmed sepsis. Several high-performing clinical signs not currently in IMCI, including prolonged capillary refill and central cyanosis, also showed strong associations with adverse outcomes.

Meaning:

This study corroborated the value of current IMCI signs and identified potential new signs that may improve the accuracy of WHO algorithms to diagnose sepsis or identify infants with high risk of mortality.

ACKNOWLEDGMENTS

Concept and design: Driker, Mathias, Fung, Lee

Design of data collection instruments: Driker, Fung

Design and conduct of search strategies: Wade

Study screening, data extraction and verification, and quality assessment: Driker, Mathias, Fung, Widyaningsih, Schmeck, Adnan, Mediratta, Kim, Hussaini, Kehoe, Hoey, Shafiq

Statistical analysis: Driker, Mathias

Interpretation of data: Driker, Mathias, Fung, Rees, North, Mediratta, Lee

Drafting of the manuscript: Driker

Critical revision of the manuscript for important intellectual content: All authors

Supervision: Lee

Driker and Lee had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis

Funding/Support:

This review was partially funded by a grant from the World Health Organization to Brigham and Women’s Hospital. Dr. Rees was supported, in part, by the US National Institutes of Health (K23HL173694). Dr. North was supported, in part, by the US National Institutes of Health (K23HD118179).

Role of Funder/Sponsor (if any):

The funders had no role in the study design or in the collection, analysis, or interpretation of the data. The funders did not write the report and had no role in the decision to submit the paper for publication.

Abbreviations

CI

Confidence interval

IMCI

Integrated management of childhood illness

LMICs

Low- and middle-income countries

NICU

Neonatal Intensive Care Unit

OR

Odds ratio

PIRD

Population, Index test, Reference Standard, Diagnosis of interest

pSBI

Possible Serious Bacterial Infections

QUADAS-2

Quality Assessment of Diagnostic Accuracy Studies-2

QUAPAS

Quality Assessment of Prognostic Accuracy Studies

RR

Risk ratio

WHO

World Health Organization

Footnotes

Conflict of Interest Disclosures (includes financial disclosures): The authors have no conflicts of interest relevant to this article to disclose.

Clinical Trial Registration (if any): None.

Protocol Registration: The protocol for this systematic review was registered at https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023431387, Prospero registration: CRD42023431387.

Data availability:

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  • 1.Perin J, Mulick A, Yeung D, et al. Global, regional, and national causes of under-5 mortality in 2000-19: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet Child Adolesc Health. 2022;6(2):106–115. doi: 10.1016/S2352-4642(21)00311-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.World Health Organization. WHO Recommendations for Management of Serious Bacterial Infections in Infants Aged 0–59 Days.; 2024. Accessed August 25, 2025. https://www.ncbi.nlm.nih.gov/books/NBK610587/ [PubMed]
  • 3.World Bank Country and Lending Groups – World Bank Data Help Desk. Accessed September 1, 2025. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups
  • 4.United Nations Inter-Agency Group for Child Mortality Estimation (UNIGME), Levels & Trends in Child Mortality: Report 2024. United Nations Children’s Fund; 2025. [Google Scholar]
  • 5.GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204–1222. doi: 10.1016/S0140-6736(20)30925-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Integrated management of the sick child. Bull World Health Organ. 1995;73(6):735–740. [PMC free article] [PubMed] [Google Scholar]
  • 7.Boschi-Pinto C, Labadie G, Dilip TR, et al. Global implementation survey of Integrated Management of Childhood Illness (IMCI): 20 years on. BMJ Open. 2018;8(7):e019079. doi: 10.1136/bmjopen-2017-019079 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Young Infants Clinical Signs Study Group. Clinical signs that predict severe illness in children under age 2 months: a multicentre study. Lancet. 2008;371(9607):135–142. doi: 10.1016/S0140-6736(08)60106-3 [DOI] [PubMed] [Google Scholar]
  • 9.World Health Organization. Integrated Management of Childhood Illness: Management of the Sick Young Infant Aged up to 2 Months: IMCI Chart Booklet. World Health Organization; 2019. [Google Scholar]
  • 10.World Health Organization. Pocket Book of Hospital Care for Children. 2nd ed.; 2013. [PubMed] [Google Scholar]
  • 11.WHO Recommendations on Maternal and Newborn Care for a Positive Postnatal Experience. World Health Organization; 2022. [PubMed] [Google Scholar]
  • 12.Fung A, Shafiq Y, Driker S, et al. Diagnostic Accuracy of Clinical Sign Algorithms to Identify Sepsis in Young Infants Aged 0 to 59 Days: A Systematic Review and Meta-analysis. Pediatrics. 2024;154(Suppl 1). doi: 10.1542/peds.2024-066588D [DOI] [PubMed] [Google Scholar]
  • 13.Seale AC, Blencowe H, Manu AA, et al. Estimates of possible severe bacterial infection in neonates in sub-Saharan Africa, south Asia, and Latin America for 2012: a systematic review and meta-analysis. Lancet Infect Dis. 2014;14(8):731–741. doi: 10.1016/S1473-3099(14)70804-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.PROSPERO. Accessed September 10, 2025. https://www.crd.york.ac.uk/PROSPERO/view/CRD42023431387
  • 15.Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions v6.2. Cochrane; 2021. Accessed October 21, 2021. https://training.cochrane.org/handbook#how-to-cite [Google Scholar]
  • 16.Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. March 29, 2021:n71. doi: 10.1136/bmj.n71 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Veritas Health Innovation. Covidence Systematic Review Software. Veritas Health Innovation; 2021. [Google Scholar]
  • 18.World Health Organization. Survive and thrive: transforming care for every small and sick newborn. 2019. Accessed August 19, 2025. https://iris.who.int/bitstream/handle/10665/326495/9789241515887-eng.pdf?sequence=1
  • 19.Ottawa Hospital Research Institute. The Newcastle–Ottawa scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. Accessed August 19, 2025. https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp
  • 20.Whiting PF, Rutjes AWS, Westwood ME, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529–536. doi: 10.7326/0003-4819-155-8-201110180-00009 [DOI] [PubMed] [Google Scholar]
  • 21.Lee J, Mulder F, Leeflang M, Wolff R, Whiting P, Bossuyt PM. QUAPAS: An Adaptation of the QUADAS-2 Tool to Assess Prognostic Accuracy Studies. Ann Intern Med. 2022;175(7):1010–1018. doi: 10.7326/M22-0276 [DOI] [PubMed] [Google Scholar]
  • 22.McPheeters ML, Kripalani S, Peterson NB, et al. Quality Improvement Interventions To Address Health Disparities. Closing the Quality Gap: Revisiting the State of the Science. Agency for Healthcare Research and Quality; 2012. [PMC free article] [PubMed] [Google Scholar]
  • 23.StataCorp. Stata Statistical Software: Release 18. StataCorp LLC; 2023. [Google Scholar]
  • 24.Meta summarize — Summarize meta-analysis data. StataCorp. [Google Scholar]
  • 25.meta forestplot — Forest plots. StataCorp. [Google Scholar]
  • 26.Deeks JJ, Higgins JP, Algman DG, Eds on behalf of the CSMG. Chapter 10: Analysing data and undertaking meta-analyses ∣Cochrane Training. 2021. Accessed October 22, 2021. https://training.cochrane.org/handbook/current/chapter-10 [Google Scholar]
  • 27.Harbord RM, Whiting P. Metandi: Meta-analysis of Diagnostic Accuracy Using Hierarchical Logistic Regression. The Stata Journal. 2009;9(2):211–229. doi: 10.1177/1536867X0900900203 [DOI] [Google Scholar]
  • 28.Nyaga VN, Arbyn M. Metadta: a Stata command for meta-analysis and meta-regression of diagnostic test accuracy data - a tutorial. Arch Public Health. 2022;80(1):95. doi: 10.1186/s13690-021-00747-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ashrafzadeh M, Shirinabadi Farahani A, Rassouli M, Shariat M, Nasiri M, Faridpor F. The Prediction of Mortality Risk in Preterm Infants Hospitalized in the Neonatal Intensive Care Unit Using SNAPPE-II Score System. J Compr Ped. 2019;10(3). doi: 10.5812/compreped.85983 [DOI] [Google Scholar]
  • 30.Bang AT, Bang RA, Reddy MH, et al. Simple clinical criteria to identify sepsis or pneumonia in neonates in the community needing treatment or referral. Pediatr Infect Dis J. 2005;24(4):335–341. doi: 10.1097/01.inf.0000157094.43609.17 [DOI] [PubMed] [Google Scholar]
  • 31.Bang AT, Reddy HM, Baitule SB, Deshmukh MD, Bang RA. The incidence of morbidities in a cohort of neonates in rural Gadchiroli, India: seasonal and temporal variation and a hypothesis about prevention. J Perinatol. 2005;25 Suppl 1:S18–28. doi: 10.1038/sj.jp.7211271 [DOI] [PubMed] [Google Scholar]
  • 32.Bekhof J, Reitsma JB, Kok JH, Van Straaten IHLM. Clinical signs to identify late-onset sepsis in preterm infants. Eur J Pediatr. 2013;172(4):501–508. doi: 10.1007/s00431-012-1910-6 [DOI] [PubMed] [Google Scholar]
  • 33.Bonadio WA, McElroy K, Jacoby PL, Smith D. Relationship of fever magnitude to rate of serious bacterial infections in infants aged 4-8 weeks. Clin Pediatr (Phila). 1991;30(8):478–480. doi: 10.1177/000992289103000803 [DOI] [PubMed] [Google Scholar]
  • 34.Bohnhorst B, Lange M, Bartels DB, Bejo L, Hoy L, Peter C. Procalcitonin and valuable clinical symptoms in the early detection of neonatal late-onset bacterial infection. Acta Paediatr. 2012;101(1):19–25. doi: 10.1111/j.1651-2227.2011.02438.x [DOI] [PubMed] [Google Scholar]
  • 35.Brady JP, Awan FB, Wafula EM, Onyango FE. Recognition of illness in very young infants by inexperienced health workers. Ann Trop Paediatr. 1993;13(4):401–407. doi: 10.1080/02724936.1993.11747680 [DOI] [PubMed] [Google Scholar]
  • 36.Coggins SA, Carr LH, Harris MC, Srinivasan L. Sepsis Huddles in the Neonatal Intensive Care Unit: A Retrospective Cohort Study of Late-onset Infection Recognition and Severity Assessment. J Pediatr. 2024;272:114117. doi: 10.1016/j.jpeds.2024.114117 [DOI] [PubMed] [Google Scholar]
  • 37.Darmstadt GL, Baqui AH, Choi Y, et al. Validation of a clinical algorithm to identify neonates with severe illness during routine household visits in rural Bangladesh. Arch Dis Child. 2011;96(12):1140–1146. doi: 10.1136/archdischild-2011-300591 [DOI] [PubMed] [Google Scholar]
  • 38.Das A, Shukla S, Rahman N, Gunzler D, Abughali N. Clinical Indicators of Late-Onset Sepsis Workup in Very Low-Birth-Weight Infants in the Neonatal Intensive Care Unit. Am J Perinatol. 2016;33(9):856–860. doi: 10.1055/s-0036-1579648 [DOI] [PubMed] [Google Scholar]
  • 39.Desalew A, Sintayehu Y, Teferi N, et al. Cause and predictors of neonatal mortality among neonates admitted to neonatal intensive care units of public hospitals in eastern Ethiopia: a facility-based prospective follow-up study. BMC Pediatr. 2020;20(1):160. doi: 10.1186/s12887-020-02051-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Diaz Alvarez M. Nuevos criterios de bajo riesgo de infección bacteriana severa en recién nacidos febriles. Rev Cubana Pediatr. 1996. [Google Scholar]
  • 41.Duke T, Oa O, Mokela D, Oswyn G, Hwaihwanje I, Hawap J. The management of sick young infants at primary health centres in a rural developing country. Arch Dis Child. 2005;90(2):200–205. doi: 10.1136/adc.2003.047951 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.English M, Ngama M, Mwalekwa L, Peshu N. Signs of illness in Kenyan infants aged less than 60 days. Bull World Health Organ. 2004;82(5):323–329. [PMC free article] [PubMed] [Google Scholar]
  • 43.Fanaroff AA. Incidence, presenting features, risk factors and significance of late onset septicemia in very low birth weight infants. The National Institute of Child Health and Human Development Neonatal Research Network . The Pediatric infectious disease journal. 1998;17(7):593–598. [DOI] [PubMed] [Google Scholar]
  • 44.Guo B-C, Chen Y-T, Chang Y-J, Chen C-Y, Lin W-Y, Wu H-P. Predictors of bacteremia in febrile infants under 3 months old in the pediatric emergency department. BMC Pediatr. 2023;23(1):444. doi: 10.1186/s12887-023-04271-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Hailemeskel HS, Tiruneh SA. Development of a nomogram for clinical risk prediction of preterm neonate death in ethiopia. Front Pediatr. 2022;10:877200. doi: 10.3389/fped.2022.877200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Hiew TM. Clinical features and haematological indices of bacterial infections in young infants . Singapore medical journal. 1992;33(2):125–130. [PubMed] [Google Scholar]
  • 47.Hofer N, Müller W, Resch B. Neonates presenting with temperature symptoms: role in the diagnosis of early onset sepsis. Pediatr Int. 2012;54(4):486–490. doi: 10.1111/j.1442-200X.2012.03570.x [DOI] [PubMed] [Google Scholar]
  • 48.Hofer N, Zacharias E, Müller W, Resch B. Performance of the definitions of the systemic inflammatory response syndrome and sepsis in neonates. J Perinat Med. 2012;40(5):587–590. doi: 10.1515/jpm-2011-0308 [DOI] [PubMed] [Google Scholar]
  • 49.Husada D, Chanthavanich P, Chotigeat U, et al. Predictive model for bacterial late-onset neonatal sepsis in a tertiary care hospital in Thailand. BMC Infect Dis. 2020;20(1):151. doi: 10.1186/s12879-020-4875-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.John B, David M, Mathias L, Elizabeth N. Risk factors and practices contributing to newborn sepsis in a rural district of Eastern Uganda, August 2013: a cross sectional study. BMC Res Notes. 2015;8(1):339. doi: 10.1186/s13104-015-1308-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Khan FA, Mullany LC, Wu LF- S, et al. Predictors of neonatal mortality: development and validation of prognostic models using prospective data from rural Bangladesh. BMJ Glob Health. 2020;5(1):e001983. doi: 10.1136/bmjgh-2019-001983 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Lee NH, Nam SK, Lee J, Jun YH. Clinical impact of admission hypothermia in very low birth weight infants: results from Korean Neonatal Network. Korean J Pediatr. 2019;62(10):386–394. doi: 10.3345/kjp.2019.00206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Lloyd LG, van Weissenbruch MM, Dramowski A, Gleeson B, Ferreyra C, Bekker A. Development and internal validation of a Neonatal Healthcare-associated infectiOn Prediction score (NeoHoP score) for very low birthweight infants in low-resource settings: a retrospective case-control study. bmjpo. 2023;7(1). doi: 10.1136/bmjpo-2023-002056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Mathur NB, Arora D. Role of TOPS (a simplified assessment of neonatal acute physiology) in predicting mortality in transported neonates. Acta Paediatr. 2007;96(2):172–175. doi: 10.1111/j.1651-2227.2007.00006.x [DOI] [PubMed] [Google Scholar]
  • 55.Mediratta RP, Amare AT, Behl R, et al. Derivation and validation of a prognostic score for neonatal mortality in Ethiopia: a case-control study. BMC Pediatr. 2020;20(1):238. doi: 10.1186/s12887-020-02107-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Modi N, Doré CJ, Saraswatula A, et al. A case definition for national and international neonatal bloodstream infection surveillance. Arch Dis Child Fetal Neonatal Ed. 2009;94(1):F8–12. doi: 10.1136/adc.2007.126458 [DOI] [PubMed] [Google Scholar]
  • 57.Mohamed SA, Kamel NR, Fouda AE, Elhawary RE, Abdelmegeid MA. Association of low vitamin D level and full-term early-onset neonatal sepsis; a case-control study. Ital J Pediatr. 2024;50(1):101. doi: 10.1186/s13052-024-01665-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Money NM, Lo YHJ, King H, et al. Predicting serious bacterial infections among hypothermic infants in the emergency department. Hosp Pediatr. 2024;14(3):153–162. doi: 10.1542/hpeds.2023-007356 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Mugalu J, Nakakeeto MK, Kiguli S, Kaddu-Mulindwa DH. Aetiology, risk factors and immediate outcome of bacteriologically confirmed neonatal septicaemia in Mulago hospital, Uganda. Afr Health Sci. 2006;6(2):120–126. doi: 10.5555/afhs.2006.6.2.120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Obaro HK, Abdulkadir B, Abdullahi S. In vitro antibiotic susceptibility of bacterial pathogens and risk factors associated with culture positive neonatal sepsis in two hospitals, Katsina metropolis, Nigeria. Af J Clin Exp Micro. 2022;23(4):378–388. doi: 10.4314/ajcem.v23i4.6 [DOI] [Google Scholar]
  • 61.Ohlin A, Björkqvist M, Montgomery SM, Schollin J. Clinical signs and CRP values associated with blood culture results in neonates evaluated for suspected sepsis. Acta Paediatr. 2010;99(11):1635–1640. doi: 10.1111/j.1651-2227.2010.01913.x [DOI] [PubMed] [Google Scholar]
  • 62.Orfanos I, Sotoca Fernandez J, Elfving K, Alfvén T, Eklund EA. Paediatric emergency departments should manage young febrile and afebrile infants the same if they have a fever before presenting. Acta Paediatr. 2022;111(10):2004–2009. doi: 10.1111/apa.16483 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Padula MA, Dewan ML, Shah SS, et al. Risk factors associated with laboratory-confirmed bloodstream infections in a tertiary neonatal intensive care unit. Pediatr Infect Dis J. 2014;33(10):1027–1032. doi: 10.1097/INF.0000000000000386 [DOI] [PubMed] [Google Scholar]
  • 64.Pathak GH, Chauhan AV, Patel PB. Correlation of TOPS scoring with immediate outcome among neonates transported to special newborn care unit: a prospective study. Int J Contemp Pediatr. 2019;6(6):2394. doi: 10.18203/2349-3291.ijcp20194705 [DOI] [Google Scholar]
  • 65.Puri D, Nisar YB, Tshefu A, et al. Prevalence of clinical signs of possible serious bacterial infection and mortality associated with them from population-based surveillance of young infants from birth to 2 months of age. PLoS ONE. 2021;16(2):e0247457. doi: 10.1371/journal.pone.0247457 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Ramgopal S, Walker LW, Vitale MA, Nowalk AJ. Factors associated with serious bacterial infections in infants ≤60 days with hypothermia in the emergency department. Am J Emerg Med. 2019;37(6):1139–1143. doi: 10.1016/j.ajem.2019.04.015 [DOI] [PubMed] [Google Scholar]
  • 67.Ramgopal S, Janofsky S, Zuckerbraun NS, et al. Risk of Serious Bacterial Infection in Infants Aged ≤60 Days Presenting to Emergency Departments with a History of Fever Only. J Pediatr. 2019;204:191–195. doi: 10.1016/j.jpeds.2018.08.043 [DOI] [PubMed] [Google Scholar]
  • 68.Ronfani L, Vilarim JN, Dragovich D, Bacalhau AF, Cattaneo A. Signs of severe bacterial infection in neonates. J Trop Pediatr. 1999;45(1):48–51. doi: 10.1093/tropej/45.1.48 [DOI] [PubMed] [Google Scholar]
  • 69.Rosenberg RE, Ahmed ASMNU, Saha SK, et al. Nosocomial sepsis risk score for preterm infants in low-resource settings. J Trop Pediatr. 2010;56(2):82–89. doi: 10.1093/tropej/fmp061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Simiyu DE. Morbidity and mortality of neonates admitted in general paediatric wards at Kenyatta National Hospital. East Afr Med J. 2003;80(12):611–616. doi: 10.4314/eamj.v80i12.8776 [DOI] [PubMed] [Google Scholar]
  • 71.Singh SA, Dutta S, Narang A. Predictive clinical scores for diagnosis of late onset neonatal septicemia. J Trop Pediatr. 2003;49(4):235–239. doi: 10.1093/tropej/49.4.235 [DOI] [PubMed] [Google Scholar]
  • 72.Singhi S, Chaudhuri M. Functional and behavioral responses as marker of illness, and outcome in infants under 2 months. Indian Pediatr. 1995;32(7):763–771. [PubMed] [Google Scholar]
  • 73.Thapa B. Neonatal sepsis as a major cause of morbidity in a tertiary center in Kathmandu . JNMA; journal of the Nepal Medical Association. 2013;52(192):549–556. [PubMed] [Google Scholar]
  • 74.Ussat M, Vogtmann C, Gebauer C, Pulzer F, Thome U, Knüpfer M. The role of elevated central-peripheral temperature difference in early detection of late-onset sepsis in preterm infants. Early Hum Dev. 2015;91(12):677–681. doi: 10.1016/j.earlhumdev.2015.09.007 [DOI] [PubMed] [Google Scholar]
  • 75.Verstraete EH, Mahieu L, d’Haese J, et al. Blood culture indications in critically ill neonates: a multicenter prospective cohort study. Eur J Pediatr. 2018;177(10):1565–1572. doi: 10.1007/s00431-018-3203-1 [DOI] [PubMed] [Google Scholar]
  • 76.Weber MW, Carlin JB, Gatchalian S, et al. Predictors of neonatal sepsis in developing countries. Pediatr Infect Dis J. 2003;22(8):711–717. doi: 10.1097/01.inf.0000078163.80807.88 [DOI] [PubMed] [Google Scholar]
  • 77.Zhang W-W, Wang S, Li Y, et al. Development and validation of a model to predict mortality risk among extremely preterm infants during the early postnatal period: a multicentre prospective cohort study. BMJ Open. 2023;13(12):e074309. doi: 10.1136/bmjopen-2023-074309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Zhang Y, Zhang C, Shu J, Zhang F. Establishment and validation of models based on clinical parameters/symptoms for diagnostic and prognostic assessment of neonatal sepsis. Iran J Pediatr. 2023;33(6). doi: 10.5812/ijp-138660 [DOI] [Google Scholar]
  • 79.Widyaningsih SA, Fung A, Driker S, Khanam R, Baqui A, Lee AC. Accuracy of Infant Clinical Signs to Predict Neonatal Mortality in Rural Bangladesh - Part II of Accuracy of Infant Clinical Signs to Predict Young Infant Mortality. ProQuest Dissertations & Theses Global: The Sciences and Engineering Collection; ProQuest One Academic; 2025. Accessed September 2, 2025. http://Available from ProQuest Dissertations & Theses Global [Google Scholar]
  • 80.Opiyo N, English M. What clinical signs best identify severe illness in young infants aged 0–59 days in developing countries? A systematic review. Arch Dis Child. 2011;96(11):1052–1059. doi: 10.1136/adc.2010.186049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.Fung A, Shafiq Y, Driker S, et al. Diagnostic accuracy of clinical sign algorithms to identify sepsis in young infants aged 0-59 days: a systematic review and meta-analysis. Pediatrics. 2024. [DOI] [PubMed] [Google Scholar]
  • 82.Kalhoff H, Kersting M, Sinningen K, Lücke T. Development of eating skills in infants and toddlers from a neuropediatric perspective. Ital J Pediatr. 2024;50(1):110. doi: 10.1186/s13052-024-01683-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.Nisar YB, Tshefu A, Longombe AL, et al. Clinical signs of possible serious infection and associated mortality among young infants presenting at first-level health facilities. PLoS ONE. 2021;16(6):e0253110. doi: 10.1371/journal.pone.0253110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.Darmstadt GL, Ahmed S, Islam MS, et al. Association of clinical signs of possible serious bacterial infections identified by community health workers with mortality of young infants in South Asia: a prospective, observational cohort study. EClinicalMedicine. 2025;80:103070. doi: 10.1016/j.eclinm.2025.103070 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 85.Hibberd PL, Hansen NI, Wang ME, et al. Trends in the incidence of possible severe bacterial infection and case fatality rates in rural communities in Sub-Saharan Africa, South Asia and Latin America, 2010-2013: a multicenter prospective cohort study. Reprod Health. 2016;13(1):65. doi: 10.1186/s12978-016-0177-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86.Schmeck N, Driker S, Yu M, et al. Diagnostic Thresholds of Fever and Hypothermia to Predict Sepsis and Mortality in Young Infants Aged 0–59 Days: A Systematic Review and Meta-analysis. [Manuscript in preparation]. Department of Pediatrics, Brigham and Women’s Hospital [Google Scholar]
  • 87.Jehan F, Qazi S. Treating Sick Young Infants With Only Fast Breathing With Oral Amoxicillin in Resource-Limited Settings: Taking the High Road? Clin Infect Dis. 2017;64(2):190–191. doi: 10.1093/cid/ciw695 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 88.Njeru CM, Ansermino JM, Macharia WM, Dunsmuir DT. Variability of respiratory rate measurements in neonates- every minute counts. BMC Pediatr. 2022;22(1):16. doi: 10.1186/s12887-021-03087-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 89.Fleming S, Gill P, Jones C, et al. The Diagnostic Value of Capillary Refill Time for Detecting Serious Illness in Children: A Systematic Review and Meta-Analysis. PLoS ONE. 2015;10(9):e0138155. doi: 10.1371/journal.pone.0138155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 90.Management of the Child with a Serious Infection or Severe Malnutrition: Guidelines for Care at the First-Referral Level in Developing Countries. World Health Organization; 2000. [Google Scholar]
  • 91.Fleming S, Gill PJ, Van den Bruel A, Thompson M. Capillary refill time in sick children: a clinical guide for general practice. Br J Gen Pract. 2016;66(652):587. doi: 10.3399/bjgp16X687925 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 92.Zensho K, Mitsui K, Ogino K. Prolonged capillary refill time as an early sign of compromised circulation in an infant with supraventricular tachycardia: A case report. Cureus. 2024;16(12):e75913. doi: 10.7759/cureus.75913 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 93.Fair FJ, Furness A, Higginbottom G, Oddie SJ, Soltani H. Systematic review of Apgar scores & cyanosis in Black, Asian, and ethnic minority infants. Pediatr Res. 2025;97(3):939–952. doi: 10.1038/s41390-024-03543-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 94.Countdown to 2030 Collaboration. Countdown to 2030: Tracking progress towards universal coverage for reproductive, maternal, newborn, and child health. Lancet. 2018;391(10129):1538–1548. doi: 10.1016/S0140-6736(18)30104-1 [DOI] [PubMed] [Google Scholar]
  • 95.Reñosa MD, Dalglish S, Bärnighausen K, McMahon S. Key challenges of health care workers in implementing the integrated management of childhood illnesses (IMCI) program: a scoping review. Glob Health Action. 2020;13(1):1732669. doi: 10.1080/16549716.2020.1732669 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material
NIHMS2178998-supplement-Supplementary_material.docx (1.5MB, docx)
Supplementary tables
NIHMS2178998-supplement-Supplementary_tables.xlsx (234.2KB, xlsx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

RESOURCES