Performance of non-contact infrared thermometer for detecting febrile children in hospital and ambulatory settings.

AIMS: To assess the performance of the non-contact infrared thermometer compared with mercury-in-glass thermometer in children; to assess the diagnostic accuracy of non-contact infrared thermometer for detecting children with fever; to compare the discomfort caused by the two procedures in children aged > one month.
BACKGROUND: Non-contact infrared thermometer is a quick and non-invasive method to measure body temperature, not requiring sterilisation or disposables. It is a candidate for temperature recording in children.
DESIGN: Prospective multicenter study.
METHODS: Body temperature readings were taken from every child consecutively admitted to the Pediatric Emergency Departments or Pediatric Clinics participating in the study. Two bilateral axillary temperature measurements using the mercury-in-glass thermometers and three mid-forehead temperature measurements using the non-contact infrared thermometer were performed.
RESULTS: Two hundred and fifty-one children were enrolled in the study. Mean body temperature obtained by mercury-in-glass thermometer and non-contact infrared thermometer was 37.18 (SD 0.96) °C and 37.30 (SD 0.92) °C, respectively (p = 0.153). Non-contact infrared thermometer clinical repeatability was 0.108 (SD 0.095) °C, similar to that of the mercury-in-glass thermometer (0.11 SD 01 °C; p = 0.517). Bias was 0.0150 (SD 0.09) °C. The proportion of outliers >1 °C was 4/251 children (1.59%). A significant correlation between temperature values obtained with the two procedures was observed (r(2) = 0.84; p < 0.0001). The limits of agreement, by the Bland and Altman method, were -0.62 (95% CI: -0.47 to -0.67) and 0.76 (95% CI: 0.61-0.91). No significant correlation was evidenced between the difference of the body temperature values recorded by the two methods and age (p = 0.226), or room temperature (p = 0.756). Calculating the receiver operating characteristic curve to determine the best threshold for axillary temperature >38.0 °C, for a non-contact infrared thermometer temperature = 37.98 °C the sensitivity was 88.7% and the specificity 89.9%. Mean distress score (on a 5-point scale) was significantly lower using the non-contact infrared thermometer than using the mercury-in-glass thermometer (1.92 SD 0.56 and 2.40 SD0.93, respectively; p < 0.0001).
CONCLUSION: Non-contact infrared thermometer showed a good performance in our study population, has the advantage of measuring body temperature in two seconds and is comfortable for children. RELEVANCE TO CLINICAL PRACTICE: Non-contact infrared thermometer may be taken into consideration when assessing body temperature in children aged > one month in hospital or ambulatory.