Recognising the seriously ill child and paediatric assessment triangle

Overview & Definition

A seriously ill child is one who may lose airway, effective breathing, circulation, neurological function or immediate safety unless the team acts promptly. You do not need a diagnosis before treating a life threat. Equally, a child who looks settled may still need urgent review because of age, history, a dangerous possible diagnosis, change from baseline or caregiver concern. One normal sign never settles the question. [4] [7] [15]

Keep four questions separate. How severe is the illness? Is an airway, breathing, circulation or neurological problem present now? Is the child getting worse? Could a dangerous diagnosis still be present despite a calm examination? PAT helps only with the first impression. Observations, history, examination, response to treatment, trend and caregiver concern answer the rest. [2] [3] [4]

[1] [15] [21]

Classification

The canonical PAT is the rapid visual and auditory assessment defined in the American Academy of Pediatrics PEPP framework. Appearance uses TICLS: Tone, Interactiveness, Consolability, Look or gaze, and Speech or cry. Work of Breathing includes visible or audible effort, abnormal sounds, retractions, nasal flaring and abnormal positioning. Circulation to Skin means pallor, mottling or cyanosis. Capillary refill, pulse, blood pressure, oxygen saturation, temperature and PEWS are measured afterwards; they are not PAT domains. [2] [3]

[2] [4]

Read the figure like this: describe exactly what you see, say which ABCDE area may be failing, and act. PAT has no numeric total. Capillary refill is not part of canonical PAT. A normal PAT means only that the three domains look normal at that moment; it is not a discharge decision. [2] [4] [7]

[1] [8] [9]

PAT combinations describe a pattern, not a diagnosis. All three domains normal means a stable first impression at that moment. Abnormal Appearance alone points toward brain or metabolic dysfunction. Work of Breathing alone suggests respiratory distress; Appearance plus Work of Breathing suggests respiratory failure. Circulation to Skin alone with preserved Appearance suggests compensated shock; Appearance plus Circulation to Skin suggests decompensated shock. All three abnormal suggests cardiopulmonary failure. Each pattern tells you where to focus ABCDE first, while history, examination, observations and response establish the cause. [2] [12]

[2] [12]

Epidemiology & Risk Factors

Do not quote one incidence for “missed deterioration.” Emergency, ward, community, retrieval and low-resource studies include different children and outcomes. The practical lesson is that every recognition tool depends on where and in whom it is used. Children with bacteraemia have had a normal PAT; caregiver concern predicted critical illness in one Australian hospital cohort; and children with medical complexity are not well represented in many studies. These findings justify senior review and reassessment, but not a universal probability or trigger. [4] [15] [16]

[15] [16] [17] [21]

Season, outbreaks, travel, contacts and exposures change which causes are likely. They do not change the order of immediate care. Say, “The exposure makes infection more likely, but ineffective breathing is the problem I must treat first.” Use the same discipline with fever, pain, crying, dehydration or anxiety. These may alter observations, but they must not automatically explain deterioration. [4] [22]

Pathophysiology

Children can maintain interaction and blood pressure while faster heart rate, increased breathing effort and vascular tone compensate. The amount of reserve and the expected observations change with age. Compensation is therefore a collection of findings, not a fixed sequence. When demand exceeds reserve, air entry and gas exchange fail, interaction falls, pulses weaken, organs malfunction and blood pressure may fall. Hypotension is worrying, but waiting for hypotension misses earlier circulatory failure. [5] [12]

Oxygen delivery depends on effective ventilation and oxygenation, blood oxygen content, cardiac output, tissue perfusion and cellular use. Failure at any point can make the child seriously ill. [12]

Cardiac output depends on heart rate and stroke volume, while vascular tone directs blood flow. Neonates and young infants have limited stroke-volume reserve, so heart rate contributes more to maintaining output. Tachycardia and vasoconstriction may preserve pressure and interaction for a time, but also increase demand. In a child who is hypoxic or deteriorating, a falling or abnormally low heart rate is an ominous sign, not evidence of recovery. This context-specific warning does not make every isolated bradycardia pre-arrest, as stated in Resuscitation Council UK 2025 PLS guidance. [5] [12]

Young infants have smaller airways, different respiratory mechanics and less reserve for their metabolic demand. Greater effort may initially maintain air movement, but fatigue can make ventilation worse even while retractions or visible effort lessen. This is why a suddenly quiet breathless infant may be sicker. There is no universal age threshold: count the rate under stated conditions and interpret it for the child’s exact age and context. At birth, lung aeration and circulatory transition require Resuscitation Council UK 2025 NLS guidance, not an older-child pathway. [5] [12]

[5] [12]

Read the figure like this: a quieter child after marked respiratory effort may be recovering or tiring. Recovery requires easier breathing with effective air entry, improving interaction and stable perfusion. Less effort with poorer air entry or interaction is failure until proved otherwise. The “inborn error” label in the figure means an inborn error of metabolism. [12] [22] [23]

[5] [12] [13]

Respiratory deterioration may start with faster breathing and more effort, then become ineffective as fatigue, limited air entry or impaired drive develops. Circulatory failure may be hypovolaemic, distributive, cardiogenic, obstructive or mixed; do not expect a fixed warm-to-cold sequence. Appearance may worsen because of hypoxia, altered ventilation, reduced perfusion, seizure, dangerous low glucose, toxin or systemic illness. Reassessment shows whether treatment helped; it does not prove one cause. [12] [13] [23]

Birth transition has different physiology and pathways from older-child resuscitation. Congenital heart disease, pulmonary hypertension, autonomic dysfunction, neuromuscular weakness and chronic respiratory support can also change the child’s usual observations and route to failure. Learn the personal baseline, communication method, device dependence and emergency plan before calling a finding “normal” or “improved.” [16] [21]

Clinical Presentation

Observe before touch unless treatment cannot wait. Describe interaction and arousability; tone and posture; cry or speech; consolability and gaze; visible and audible breathing effort; symmetry and apparent air entry; pallor, mottling or cyanosis; and caregiver-child interaction. Replace “looks sick” with what you can see and hear. If any finding is worrying, call for help and start ABCDE while someone else obtains the focused history. [1] [2] [22]

[15] [16] [17]

For the neonate and young infant, the named Victorian source is the Royal Children’s Hospital Melbourne seriously unwell neonate and young infant guideline. It supports prompt senior review, risk-directed assessment and bedside glucose, and warns that serious illness may be afebrile or non-specific. The caregiver’s account of change from baseline remains important. [15] [16]

Respiratory distress means increased effort while breathing remains effective. Respiratory failure means oxygenation or ventilation is ineffective and Appearance is often abnormal. A quieter child may be improving, but less effort with poor air entry, reduced interaction or worse perfusion suggests fatigue. Pulse oximetry can overestimate oxygenation, including in children with darker skin, and it does not measure ventilation. Trust the number only when the signal is credible and agrees with the child. [2] [11] [22]

Assess circulation as a whole: pulse rate and quality, skin temperature and colour, capillary-refill technique and result, mental state, urine output, blood pressure, device function and trend. Normal blood pressure or capillary refill alone does not exclude shock. [7] [12]

Serious neurological illness may appear as irritability, poor feeding, reduced interaction, abnormal tone or posture, seizure or altered behaviour rather than a verbal complaint. [23]

Differential Diagnosis

Build the differential in this order: what can kill the child now, what is most likely, what is reversible, and what is most harmful to miss. PAT tells you which part of the child may be failing. It does not tell you why. Keep the important cause groups open until focused history, examination, targeted tests and response to treatment make them less likely. [1] [12] [23]

[12] [20] [22] [23]

A clear synthesis sounds like this: “This child has abnormal Appearance and Work of Breathing, so I am treating respiratory failure. Air entry and interaction are worsening. Airway, lower-airway, parenchymal, impaired-drive, cardiac and toxic causes remain possible. I will support breathing, check reversible causes, reassess the effect and call for senior airway or critical-care help before the exact diagnosis is settled.” [1] [11] [22]

Clinical & Bedside Assessment

Observe, engage, then examine

Keep the child with the caregiver when possible. Observe breathing and interaction before handling, use play or feeding when safe, and examine the least distressing parts first. Distress can change rate, effort and behaviour and can worsen an unstable airway. These steps improve the examination; they must not delay ABCDE in an unstable child. [1] [22]

[5] [7] [11] [20] [23]

D — Disability: exact AVPU and RCH age-adapted pGCS

Use AVPU for a rapid screen: A = Alert; V = Responds to voice; P = Responds to pain; U = Unresponsive. The Royal Children’s Hospital Melbourne altered-conscious-state guideline directs any V, P or U finding to formal Glasgow Coma Scale assessment. Record AVPU and reassess after immediate ABCDE care. Also perform age-adapted pGCS when baseline concern persists despite an apparently Alert screen. Never convert an AVPU letter into an assumed pGCS number. [24] [25]

This is the exact RCH under-4 and 4-or-older implementation, not a universal developmental boundary. Add E, V and M for a total out of 15, but record each component, the total, time and trend because the same total can hide different patterns. Document the child’s usual developmental, communication and neurological baseline. If a component cannot be tested, record why rather than scoring “no response.” Use the locally endorsed chart if it differs. AVPU and pGCS do not replace pupils, posture, tone, movement, glucose assessment or the rest of the neurological examination. [24] [25]

Capillary refill: methods and cut-offs must remain separate

Capillary refill is not a PAT domain. Use it later during C of ABCDE. Record the site, compression time, ambient conditions and local chart, then interpret it with pulse, skin, mental state, urine output and blood pressure. A normal result cannot rule out serious illness. [6] [7]

[6] [7]

Focused history while stabilisation proceeds

Start with the caregiver: “What is different from usual, and what worries you most?” Ask about onset, pace, breathing, feeding, urine output, interaction, sleep and mobility. Then ask about pain, temperature change, rash, collapse, seizure, exposure, trauma, ingestion, medicines, allergies, immunisation and recent healthcare. When relevant, ask privately about pregnancy, substance use, eating restriction and self-harm. Use the child, ambulance staff, records, personal plans and devices as additional sources. Do this in parallel without interrupting life-saving care. [15] [16] [18]

For a child with complex needs, first establish usual communication, behaviour, mobility and pain expression. Then establish usual heart rate, respiratory rate, saturation, respiratory support, feeding, urine output and device function. Ask about recent changes, the emergency plan, limits of treatment and what has worked before. A chronically abnormal observation is not automatically safe; compare it with the personal baseline and trend. [16] [21]

Measurement and synthesis

Heart and respiratory rates change continuously with age. Published ranges differ because studies of healthy children, low-acuity emergency patients and hospitalised children answer different questions. The Royal Children’s Hospital Melbourne acceptable ranges for physiological variables are ranges for unwell children in that setting, not universal normal values. Say whether a number is a reference range, local escalation trigger, disease-risk threshold, treatment target or the child’s baseline. Do not merge them into one universal table. [5]

Use the correct pulse-oximetry probe and site. Check that the waveform or signal is credible, the displayed pulse matches the child and perfusion is adequate. Pulse oximetry may overestimate oxygenation in darker skin. Temperature, lighting and pigmentation also affect visual recognition of pallor, cyanosis and mottling, so examine in good light and combine several sites and signs. [11]

[1] [2] [12]

After each assessment or treatment, say and document: Is the child stable now? Which ABCDE problem is most urgent? Are they improving or worsening? What differs from baseline? Which reversible causes remain? What did the last action change? Can this service provide the next likely support? Where should the child go? Record the time, conditions and who raised concern. [15] [18] [21]

Investigations

Stabilise first, then order a test for a named question. A useful test finds an immediately reversible cause, measures failing physiology, separates urgent pathways or changes the destination. A normal panel cannot overrule a child who is worsening or remains high risk. [10] [11] [23]

[10] [11] [22] [23] [4] [10] [13] [23]

Glucose: a reversible threat without a universal threshold

Check bedside glucose promptly when the child has altered consciousness, seizure, shock, poor intake, or possible toxic or metabolic illness. If feasible, confirm an unexpected point-of-care result with a blood gas or laboratory sample because meters may be inaccurate at low levels. Do not delay correction of a clinically dangerous low. Use the current age- and context-specific pathway, recheck the glucose and escalate if it does not correct. The named Victorian source is the Royal Children’s Hospital Melbourne hypoglycaemia guideline. This page deliberately gives no universal threshold, agent, dose or infusion. [23]

Imaging and procedures

In typical bronchiolitis, chest radiography rarely changed the clinical diagnosis and was followed by more antibiotic use. NICE NG9 bronchiolitis guidance also advises against routine chest radiography because changes can mimic pneumonia and should not determine antibiotic treatment. This is a bronchiolitis-specific warning, not a ban on targeted imaging in another child. Lumbar puncture, transport-dependent imaging or a sedated procedure is premature when instability makes the procedure or movement unsafe; stabilise and seek senior or specialist input first. [10] [22]

Management — Resuscitation

A worrying doorway assessment starts age-adapted ABCDE immediately. Say that you are concerned, call the appropriate team, name a leader, assign roles, use closed-loop communication, prepare age- and weight-appropriate equipment, and monitor continuously where possible. Treat each problem when found, then reassess from A. History, sampling and cause-specific treatment can proceed in parallel only when they do not delay resuscitation. [1] [13] [18]

[13] [15] [18] [21]

Read the figure like this: the child, caregiver and personal baseline stay at the centre. The loop is constant, but exact rules depend on the disease, phase, location and available support. [13] [21]

Check glucose when altered consciousness, seizure, shock, poor intake or toxic or metabolic concern makes it relevant; correct a dangerous low through the local pathway and recheck. Under RCUK 2025 PLS guidance, an ongoing convulsive seizure activates status treatment at 5 minutes. Fluid and vasoactive decisions depend on shock type, available rescue and the local pathway. Resuscitation at birth follows RCUK 2025 NLS guidance. [13] [23]

[12] [13] [18] [23]

Emergency working weight when immediate weighing is unsafe

Use a measured weight if it can be obtained immediately without delaying life-saving care. Otherwise announce and document one temporary working weight in kilograms, including its source, method and time. [26]

Follow the local protocol: a recent reliable measured weight or credible caregiver estimate, or the service’s trained length-and-habitus method. RCUK 2025 PLS guidance uses parent or person-with-parental-responsibility report before a habitus-corrected length method in its UK hierarchy. ANZCOR Guideline 12.2 directs clinicians to local charts. Published methods vary in accuracy and are not interchangeable. [26]

Use the declared working weight with the current local paediatric cognitive aid for medicines, fluid, energy and weight-linked equipment. Still check the drug-specific body-size measure and maximum, and confirm equipment fit and function. [26]

Caregiver estimates may be stale, guessed or in the wrong units. Length-and-habitus tools depend on the population, body habitus, device version, technique and training. Total body weight is not correct for every drug or device. Age-only equations are a last local fallback, not an improvised universal formula. Re-weigh at the first safe opportunity, announce the measured value, replace the estimate and recalculate ongoing treatment. [26]

Oxygen: phase, population and jurisdiction must stay attached

RCUK 2025 PLS guidance states that respiratory, circulatory or neurological failure receives 100% oxygen initially, followed by titration. For a previously healthy child it gives a 94–98% target, using the lowest inspired oxygen fraction that achieves at least 94%. These are UK acute-rescue facts. They are not universal targets for post-resuscitation care, ventilation, chronic respiratory support or cyanotic congenital heart disease. Confirm a credible oximeter signal and follow the personal emergency plan and local pathway. [11]

Convulsive seizure: activation begins at 5 minutes

Time the seizure, protect airway and breathing, monitor, check glucose and prevent injury. RCUK 2025 PLS guidance defines a seizure continuing for at least 5 minutes as status epilepticus requiring first-line benzodiazepine treatment. Activate status treatment at 5 minutes; do not wait for two doses to fail. [23]

Circulatory support: source-specific rules are not interchangeable

First decide whether the child has shock using the whole circulation assessment. Consider the likely shock type, whether hypotension is present, whether malnutrition or fluid intolerance increases risk, and whether ventilation, critical care and vasoactive rescue are available. Decide what improvement you expect before giving fluid, reassess after each aliquot, and stop for overload or no benefit. A maximum first-hour volume is a ceiling, not a target. [12] [13] [14]

[13] [14]

FEAST found increased early mortality with saline or albumin bolus versus no bolus in its African severe-febrile-illness population. It does not prove that all boluses are harmful everywhere; it proves that a high-resource algorithm cannot be exported without regard to population and available rescue resources. If vasoactive support may be needed, activate critical-care or retrieval help in parallel. There is no audit-approved universal timing or first agent for this page; selection, concentration, preparation, access and titration belong to the dedicated shock pathway and local drug chart. [13] [14]

NLS and PLS apply at different times

Use Newborn Life Support (NLS) for resuscitation at birth and support of transition. RCUK 2025 NLS guidance is the official source for that period. During the first hospital stay, local policy defines the NLS-to-PLS boundary while experienced help is called. RCUK 2025 PLS guidance is the corresponding older-child source. [12]

Management — Definitive & Stepwise

After immediate resuscitation, move into the pathway for the likely cause without stopping reassessment. This page does not reproduce full shock, ventilation, status, antimicrobial, toxicology or glucose algorithms. Treat the most reversible dangerous problem first while keeping other serious causes open. [12] [13] [23]

[13] [18] [21]

Improvement must make sense across several findings. An open airway must also remain open. Less work of breathing is reassuring only when air entry and interaction improve. Better blood pressure does not prove adequate perfusion. A corrected glucose needs a documented recheck and a plan for the cause. A low PEWS does not cancel concern. Reassess after each action and meaningful change rather than waiting for a universal interval. [7] [8] [12] [23]

Early PICU and retrieval consultation

Call PICU or retrieval when the child may need support your service cannot reliably provide, not after every local option has failed. Discuss the likely support, present monitoring and access, equipment and staff limits, transport time and weather, safest destination, escort, expected deterioration and what to do if transfer is delayed. Exact referral thresholds, escort arrangements, transport mode and accepting service remain local or regional. [21]

[15] [17] [18] [21]

Keep the caregiver present when safe and assign someone to explain what is wrong, what is happening next and what remains uncertain. Speak directly to the child at their developmental level. Use a professional interpreter or communication aid when needed and check understanding. The caregiver’s knowledge of the child is clinical information. [15] [16] [17]

Specific Subtypes & Scenarios

The recognition loop stays the same, but age, presentation, baseline and setting change what you look for and how early you escalate. In each scenario: stabilise first, identify what is failing, activate the correct pathway and state which facts are specific to that source or setting. [1] [12] [16]

[7] [12] [20] [23] [3] [15] [16] [19] [21]

Complications & Pitfalls

Most harm begins when assessment stops too early. A normal number, convenient explanation or brief response is accepted as the answer, so the child is not reassessed. Prevent this by recording the current problem, serious alternatives, intended result, actual response and next escalation action. [7] [8] [18]

[7] [8] [10] [14] [15] [18] [20]

At handover, watch for anchoring on the previous diagnosis, choosing tests only to confirm it, carrying forward the referral label and hierarchy that silences bedside concern. PAT followed by ABCDE and reassessment helps only if the team is willing to revise its view when the child or trend does not fit. [8] [18]

After stabilisation, look actively for harm from treatment. Check airway or access complications, fluid overload, heat loss, repeated sampling, medication or device error, and pressure or restraint injury. Also look for deterioration during movement. The fact that the child did not arrest does not prove an intervention or transfer was harmless. [14] [21]

Prognosis & Disposition

Disposition follows the trend and response, not one reassuring observation. Consider the child’s age, reserve, underlying disease and the duration and severity of organ dysfunction. Then consider reversibility, treatment harm, local support, transport risk and whether the family can carry out the plan. There is no universal observation time or PICU threshold on this page. [12] [19] [21]

[15] [19] [21]

A useful safety net says what change to watch for, how urgently to act, and exactly where or how to get help. It also states when review will occur, who owns pending results and what to do if access is difficult. NICE fever-in-under-5s guidance provides an England-and-Wales example. “Return if worried” alone is not enough. No leaflet, video, call-back interval or follow-up time is best for every child. Check understanding with teach-back or another method and document it. [19]

[16] [17] [19]

After a critical event, explain what happened and what remains uncertain. Invite questions and caregiver observations, and consider debriefing for the child, family and team when appropriate. [18]

Recurrent presentation or unexplained deterioration needs senior diagnostic review and may need safeguarding reassessment. [20]

For young people moving to adult services, transfer the baseline, devices, emergency plan, capacity and caregiver roles instead of rebuilding them during the next crisis. [16]

Special Populations

Adapt how you observe, communicate and examine, but do not lower the standard for recognising deterioration. Evidence is sparse for recognition tools in disability, neurodiversity, technology dependence, telehealth and many low-resource settings. Present these adaptations as safer communication and assessment practices, not as proven improvements in diagnostic accuracy. [3] [16]

[11] [16] [17] [19] [20]

Professional interpreters improve included outcomes compared with ad hoc or no interpretation, although the evidence covers limited populations and delivery methods. Speak directly to the child in developmentally appropriate language, state the immediate concern and next action, acknowledge uncertainty and check understanding. An interpreter or communication aid is a safety measure, not a courtesy. [17]

Safeguarding runs in parallel with medical care, but immediate stabilisation and safety come first. Ask only necessary open, non-leading questions. Explain confidentiality and information sharing, document objective findings and spontaneous words accurately, preserve evidence when safe, and use the local safeguarding pathway. Reporting thresholds, consent rules and agencies differ by Australian state or territory, UK nation, US state and Canadian province or territory. [20]

Evidence, Guidelines & Regional Differences

What the evidence can and cannot support

PAT gives a consistent rapid first impression, and trained staff can use it. Its reliability and validity still vary with the domain, rater, training, setting, exclusions and outcome studied. There is no justified single pooled sensitivity, specificity, area under the curve or kappa for all settings. Neonates, older adolescents, remote and low-resource care, and children with complex baselines remain under-represented. A normal PAT cannot exclude serious infection or evolving disease. [2] [3] [4]

PEWS tools differ and depend on the chart, complete observations, training, staffing, response system and governance. Positive predictive value may be low. The EPOCH cluster randomised trial found no reduction in all-cause hospital mortality after BedsidePEWS implementation. Teach the full recognition-and-response system, not the score alone. Clinician or caregiver concern can justify escalation despite the number. [8] [9] [15]

There is no justified universal vital-sign range, PEWS chart, capillary-refill method, oxygen target, glucose threshold, fluid strategy, vasoactive timing, retrieval trigger, safeguarding route, handover mnemonic, observation interval or safety-net medium. Attach each rule to its population, purpose, conditions, jurisdiction and available support. Where evidence is weak, say so rather than inventing precision. [5] [6] [8] [13] [19] [21] [23]

[9] [14]

Caregiver concern has a strong recent signal, but it comes from one Australian hospital cohort and does not establish a universal likelihood ratio or response tier. Evidence supports professional interpretation but remains limited. Adaptations for complex children and disability are clinically necessary, although direct diagnostic-accuracy evidence is sparse. Pulse-oximetry disparity across skin colour is credible; this page offers no correction factor. [11] [15] [16] [17]

Jurisdictional implementation

Board and assessment distinctions

RACP current PREP curriculum candidates in 2026 must keep those learning objectives separate from the renewed curriculum for new first-year trainees from 2027. [1]

RCPCH Progress+ outcomes are curriculum expectations; MRCPCH FOP, TAS and AKP are theory classifications, while Clinical History, Communication, Video and short clinical stations are assessment formats. [15] [17]

The 2024 public MRCPCH face-to-face normal circuit describes two Communication and two Video stations of 9 minutes each; do not invent marks, pass standards or confidential station content. The clinical reason to preserve these distinctions is that concern, communication access and family escalation are evaluated in different instruments, not interchangeable scores. [15] [17]

The RACP General Paediatrics specialty-development page is the implementation source for that 2026-to-2027 transition; the RCPCH Generic Syllabus for Specialty Paediatric Training provides the specialty-level board mapping used alongside Progress+. [18]

The ABP General Pediatrics Content Outline classifies written examinations; ABP Core EPAs are workplace activities. [18]

ACGME Pediatrics Milestones and the 2026 Program Requirements govern residency development and accreditation, not ABP certification. Milestone Level 4 is a goal, not an automatic graduation or certification requirement. [18]

RCPSC Transition to Discipline EPA 1 and Foundations EPA 1 separate the basic history, examination and presentation task from deterioration recognition and initial stabilisation. Structured handover and transfer planning based on available support remain clinically important across these frameworks. [18] [21]

[15] [18] [21]

Exam Pearls

[1] [6] [13] [18] [19] [23]

Viva: the tachypnoeic child becomes quiet

Do not say “improved” until breathing effectiveness is reassessed. Recheck Appearance and full ABCDE, including air entry, work, respiratory rate under stated conditions, credible saturation, interaction, perfusion and glucose when relevant. Less effort with poorer air entry or interaction means fatigue and impending failure. Call airway or critical-care help, support breathing through the active pathway, state the improvement you expect and reassess from A. [11] [12] [22] [23]

Viva: concern exceeds the PEWS

Say: “The score is one input. I am concerned because interaction, breathing or perfusion and the direction of change differ from baseline, and the caregiver reports deterioration. I am escalating now through the clinician- or family-concern route while I repeat ABCDE and verify the observations.” A low score cannot cancel a high-risk presentation, worsening trend or baseline exception. [8] [9] [15]

Structured oral: rural retrieval

State the current ABCDE problem and treatment already under way. Ask what monitoring, airway, vascular-access, staff and transport resources are available. Call retrieval early, agree the destination, treatment to continue, expected response, deterioration plan and handover, and explain uncertainty and transport delay to the family. Do not invent a universal threshold or wait until all local options have failed. [18] [21]

References

1. [1]Dieckmann, Ronald A The pediatric assessment triangle: a novel approach for the rapid evaluation of children. Pediatric emergency care, 2010.PMID 20386420
2. [2]Horeczko, Timothy The Pediatric Assessment Triangle: accuracy of its application by nurses in the triage of children. Journal of emergency nursing, 2013.PMID 22831826
3. [3]Tørisen, Tore A G Emergency pediatric patients and use of the pediatric assessment triangle tool (PAT): a scoping review. BMC emergency medicine, 2024.PMID 39227775
4. [4]Gomez, B Bacteremia in previously healthy children in emergency departments: clinical and microbiological characteristics and outcome. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology, 2015.PMID 25252630
5. [5]Fleming, Susannah Normal ranges of heart rate and respiratory rate in children from birth to 18 years of age: a systematic review of observational studies. Lancet (London, England), 2011.PMID 21411136
6. [6]Fleming, Susannah Validity and reliability of measurement of capillary refill time in children: a systematic review. Archives of disease in childhood, 2015.PMID 25260515
7. [7]Fleming, Susannah The Diagnostic Value of Capillary Refill Time for Detecting Serious Illness in Children: A Systematic Review and Meta-Analysis. PloS one, 2015.PMID 26375953
8. [8]Trubey, Rob Validity and effectiveness of paediatric early warning systems and track and trigger tools for identifying and reducing clinical deterioration in hospitalised children: a systematic review. BMJ open, 2019.PMID 31061010
9. [9]Parshuram, Christopher S Effect of a Pediatric Early Warning System on All-Cause Mortality in Hospitalized Pediatric Patients: The EPOCH Randomized Clinical Trial. JAMA, 2018.PMID 29486493
10. [10]Schuh, Suzanne Evaluation of the utility of radiography in acute bronchiolitis. The Journal of pediatrics, 2007.PMID 17382126
11. [11]Sharma, Megha Racial and skin color mediated disparities in pulse oximetry in infants and young children. Paediatric respiratory reviews, 2024.PMID 38233229
12. [12]Bjorklund, Ashley Pediatric Shock Review. Pediatrics in review, 2023.PMID 37777656
13. [13]Weiss, Scott L Surviving Sepsis Campaign International Guidelines for the Management of Sepsis and Septic Shock in Children 2026. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies, 2026.PMID 41869844
14. [14]Maitland, Kathryn Mortality after fluid bolus in African children with severe infection. The New England journal of medicine, 2011.PMID 21615299
15. [15]Mills, Erin Association between caregiver concern for clinical deterioration and critical illness in children presenting to hospital: a prospective cohort study. The Lancet. Child & adolescent health, 2025.PMID 40451224
16. [16]Kuo, Dennis Z Recognition and Management of Medical Complexity. Pediatrics, 2016.PMID 27940731
17. [17]Boylen, Susan Impact of professional interpreters on outcomes for hospitalized children from migrant and refugee families with limited English proficiency: a systematic review. JBI evidence synthesis, 2020.PMID 32813387
18. [18]Starmer, Amy J Changes in medical errors after implementation of a handoff program. The New England journal of medicine, 2014.PMID 25372088
19. [19]Burvenich, Ruben Effectiveness of safety-netting approaches for acutely ill children: a network meta-analysis. The British journal of general practice : the journal of the Royal College of General Practitioners, 2025.PMID 39117428
20. [20]Bragança-Souza, Kátia Kely Health Professionals: Identifying and Reporting Child Physical Abuse-a Scoping Review. Trauma, violence & abuse, 2024.PMID 36747372
21. [21]Chaichotjinda, Krittiya Assessment of interhospital transport care for pediatric patients. Clinical and experimental pediatrics, 2020.PMID 31477679
22. [22]Shah, Sonal N Does This Child Have Pneumonia?: The Rational Clinical Examination Systematic Review. JAMA, 2017.PMID 28763554
23. [23]Faustino, E Vincent S Hypoglycemia in critically ill children. Journal of diabetes science and technology, 2012.PMID 22401322
24. [24]Hoffmann, Florian Comparison of the AVPU Scale and the Pediatric GCS in Prehospital Setting. Prehospital emergency care, 2016.PMID 26954262
25. [25]Ramgopal, Sriram Comparing AVPU and Glasgow Coma Scales Among Children Seen by Emergency Medical Services. Pediatrics, 2024.PMID 38993159
26. [26]Young, Kelly D Weight Estimation Methods in Children: A Systematic Review. Annals of emergency medicine, 2016.PMID 27105839
27. [27]McIntyre, John Safety and efficacy of buccal midazolam versus rectal diazepam for emergency treatment of seizures in children: a randomised controlled trial. Lancet (London, England), 2005.PMID 16023510