The average age of participant was 4.72 months (SD 3.12, range 1-11 months). The range of temperatures varied from 98.3°F-103°F for the rectal measurement and 97.7°F-103.2°F for the adjusted pacifier measurement. Using a paired samples t-test, difference scores (rectal temperature-[6-minute pacifier temperature + 0.5oF]) were calculated for each subject (see Table 1). The mean difference was 0.012°F (SD 0.777, range was -2.0 to +1.2). The 95% confidence interval for the difference scores was -0.309 to +0.333. There was no statistically significant (p=0.939) difference between the rectal and adjusted pacifier measurements (see Table 2).
The correlation between the rectal temperature and adjusted pacifier temperature was r=0.772 (see Table 2). The correlation was statistically significant. To correct for the potential bias inherent with a small sample size, an adjusted correlation coefficient was also computed. The adjusted correlation between rectal and pacifier temperature was radj=0.761. The close relationship between r and radj indicates a relatively unbiased correlation coefficient. There was also a strong correlation between the 3- and 6-minute pacifier thermometer measurements (r=0.913). Of the 25 children, 4 (16%) were febrile. Of these febrile subjects, final rectal and pacifier temperatures were very highly correlated (r=0.934). In this same group of febrile subjects, 3- and 6-minute pacifier readings were also highly correlated (r=1.00).
In regard to clinical reliability and utility, 88% of the rectal and adjusted pacifier temperature measurements were within 1.0°F. For 13 of 25 subjects (52%), the adjusted pacifier temperature was actually higher than the rectal temperature measurement by 0.2°F to 2.0°F (mean=0.6°F). In 2 patients, there was no difference between adjusted pacifier and rectal temperatures. The remaining infants had a higher rectal temperature than adjusted pacifier temperature. Specificity calculations indicated 86% of afebrile participants (n=18 out of 21) were correctly identified as afebrile. Sensitivity calculations were also performed; 3 out of 4 febrile subjects (75%) were correctly identified as febrile (100.4°F or higher, rectally). Table 3 outlines the concordance between rectal and adjusted pacifier temperatures based on febrile or afebrile status. Interestingly, 12% (n=3) of the subjects were determined to be febrile by the adjusted pacifier temperature and not the rectal temperature.
Pediatr Nurs. 2006;32(5):413-418. © 2006 Jannetti Publications, Inc.
Cite this: Accuracy of Pacifier Thermometers in Young Children - Medscape - Sep 01, 2006.