AIM: The risk of serious illness in febrile infants (<60 days old) is high, and so fever often warrants aggressive management. Infrared thermometers are unreliable in young infants despite their ubiquity. We aim to describe the: (i) frequency of infrared thermometer usage; (ii) progression to documented fever in the emergency department (ED) and (iii) rate of serious illness (meningitis, urinary tract infection and bacteremia). METHODS: In this single-centre retrospective chart review at The Royal Children's Hospital, Melbourne, we audited medical records of infants (<60 days old) presenting to the ED with pre-hospital fever on history over a 12-month period. We described the type of thermometer used at home (tympanic or forehead, 'infrared' vs. axillary or rectal, 'direct') correlated to peak temperature in ED, investigations, treatment and diagnosis. The primary outcome was subsequent fever in ED. RESULTS: Of 159 infants, two of three had infrared temperature measurement at home. Fifty-one (32.1%) developed fever in ED (direct 28/54, 52% vs. infrared 23/105, 22% RR 2.36 (95% CI 1.52-3.69)). Investigations (75%) and admission (60%) were common. Pre-hospital fever alone was less likely to be associated with serious illness, with fever in ED a much stronger predictor. CONCLUSIONS: In young infants, infrared thermometer use is common and less likely to predict subsequent fever. Twenty-two percent of infants with fever via infrared measurement had fever in ED. History of fever without confirmation is less likely to signal serious illness. Education to public and health-care providers is required to avoid usage of infrared devices in this population.