[Investigating relationships among five temperature measurement sites in newborns]

Environmental conditions can easily affect the body temperature of newborns because their central nervous system, which helps maintain body temperature, has not yet fully developed. Monitoring changes in newborn body temperature can help prevent harm caused by abnormally high and low body temperatures. Taking rectal temperature readings is the most reliable way to obtain core temperature. However, because inserting a thermometer into a newborn's rectum may cause discomfort and present health risks, this temperature-taking approach is no longer used in clinical settings. The axillary temperature readings now used, including ear and back temperature, have not been empirically verified to accurately reflect core body temperature. Convenient-to-use electronic thermometers have replaced traditional mercury thermometers in the clinic. However, nearly all research done until recent years has employed mercury thermometers for temperature measurement, there is a paucity of research employing electronic thermometers.

The purpose of this study was to investigate relationships among five temperature measurement sites (including tympanic temperature, rectal temperature, axillary temperature, and back temperature) in newborns.

This study used a descriptive-correlation design, employed convenience sampling, and collected data from August 2010 to February 2011 at a registered postpartum nursing home in Taipei. Thirty-three babies aged from 1 to 28 days old were recruited. MT-200 electronic thermometers were utilized to take back, axillary, and rectal temperature readings and an IRT4520 ear thermometer was used to take ear temperature readings. Collected data were analyzed using SPSS18.0 statistical software for descriptive statistics, Pearson correlation, generalized estimation equations (GEE), and intraclass correlation coefficient (ICC).

We identified the least difference between right tympanic temperature and rectal temperature' a significant relationship between left tympanic temperature and rectal temperature, and axillary temperature as the most stable of the 5 body-temperature reading methods, with no outliers. There were no gender differences among the five body temperatures. Results revealed the ICC of the five temperature measurement sites to be low during both morning and evening measurements.

Current research suggests that tympanic temperature can replace rectal temperature readings. Findings may provide a reference to clinical nursing staffs and parents to measure newborn body temperature.