Fever is the commonest presenting complaint in paediatric primary care but 1–3% of febrile children have a serious bacterial infection (SBI). The NICE traffic light system (NG143) stratifies risk; red features mandate immediate action before any other assessment.
Meningococcal disease has a case fatality of ~10% — pre-hospital IM benzylpenicillin (300mg <1yr, 600mg 1–9yr, 1.2g ≥10yr) reduces mortality and should not delay transfer.
Infants under 3 months have immature immune responses; even low-grade fever carries a 10–15% risk of SBI. Febrile infants <28 days should always be admitted.
The NICE traffic light system has a sensitivity of ~90% for SBI when red features are present. However specificity is lower — amber features require clinical judgement. No single feature is pathognomonic; the gestalt of the unwell child matters.
Temperature height alone is a poor predictor of SBI. A child with 39°C who is alert, playful and well-perfused is far lower risk than a lethargic child with 38.2°C.
Age-based thresholds reflect immuno-developmental physiology. Neonates lack maternal IgG for many pathogens and have immature neutrophil function. The NICE NG143 thresholds are derived from prospective cohort studies of 5000+ febrile children.
UTI is the most common SBI in children 1–24 months and is easily missed — it causes no localising symptoms in infants. A clean-catch urine dipstick (leucocyte esterase + nitrites) has 88% sensitivity; send MSU if positive or clinical suspicion.
Up to 20% of febrile children with a clinical diagnosis of "viral URTI" have a concomitant UTI. Systematic examination changes management in approximately 30% of cases where a source is found.
The "general impression" — how sick does this child look? — remains the highest-yield clinical decision tool and is what parents mean when they say "they just don't seem right." Trust parental concern: it correlates with illness severity.
CRP has a sensitivity of ~80% for SBI at a threshold of 80mg/L but misses ~20% of cases. It should be used to support, not replace, clinical judgement. A normal CRP does not rule out SBI in the first 12 hours of illness.
Urine testing is the most impactful investigation in primary care — UTI is the most common SBI, missed in 50% of infants because symptoms are non-specific. A clean-catch dipstick takes 2 minutes and changes management.
The decision to admit is the most important clinical decision in paediatric fever. Under-referring risks missing SBI; over-referring exposes children to nosocomial infection, unnecessary procedures, and parental anxiety. The NICE traffic light system guides but does not replace clinical judgement.
Parental concern is a validated risk factor for SBI independent of clinical findings. Studies show parents correctly identify deterioration before clinicians in 70% of cases.
Antipyretics reduce fever by 1–2°C and improve child comfort but do NOT reduce risk of febrile seizure recurrence (Cochrane 2017). Do not wake a sleeping child to give antipyretics. Fever itself is not harmful — it is part of the immune response.
Antibiotic prescribing for viral URTI is the commonest source of unnecessary antibiotic use in paediatrics and drives resistance. The RCGP TARGET Toolkit supports delayed prescription strategies. NNT to prevent one case of otitis media with antibiotics for URTI = 6000.
For UTI: trimethoprim resistance is rising (>30% in some areas). Check local antibiogram. Send MSU before treatment — culture results guide switch if no improvement at 48hrs.
Dehydration is the primary complication of fever in children and is prevented by consistent oral fluid intake rather than temperature management. Signs of dehydration (dry mucous membranes, sunken eyes, reduced urine output) should trigger urgent reassessment.
Parental anxiety is the primary driver of re-consultation for fever. Studies show that written safety-net advice reduces emergency re-attendance by 40% compared to verbal advice alone. The NICE "fever in under 5s" patient information leaflet is freely available and evidence-based.
Safety-netting reduces medicolegal risk and improves outcomes. NICE NG143 specifies that all children discharged from primary care with fever must receive written or verbal safety-net advice. Document that this was given.
Renal tract imaging after UTI detects vesicoureteric reflux (present in 30% of children with febrile UTI) and renal scarring. Missed reflux leads to recurrent UTIs and chronic kidney disease. NICE recommends USS within 6 weeks for all febrile UTIs in children under 6 months and under 3 years with unusual organisms or recurrence.