The purpose of this study was to assess the accuracy of the TempTraq thermometer as compared to a gold standard (Core temp measured by Pulmonary Artery Catheter) and to assess the precision of the TempTraq thermometer over repeated measures.
Sandra Siedlecki, Robert Butler
Learning Objectives: To be useful, a thermometer must demonstrate precision and accuracy. One alternative to the use of invasive temperature monitoring that may be more accurate and reliable than either chemical or electronic thermometers is the TempTraq® Thermometer. The TempTraq® thermometer is a thin, single-use, battery-powered skin patch that monitors and records skin temperature. Previous studies have demonstrated that it is reliable and accurate for monitoring temperature in children and in healthy adults. However, it has not been tested in ill adult patients. The purpose of this study was to assess the accuracy of the TempTraq thermometer as compared to a gold standard (Core temp measured by Pulmonary Artery Catheter) and to assess the precision of the TempTraq thermometer over repeated measures.
Methods: A repeated measures within-group comparative design was used to address the specific aims of this study. A sample of 40 was needed (Power .80) to identify a difference of 0.1 or less. To account for attrition due to repeated nature of the design, 60 patients were enrolled in this trial.
Results: Accuracy was assessed by testing the differences using a two sided paired t-test. The mean of the differences was significantly different from zero (p < .0001) and the mean of the differences was in accordance with the Bland-Altman test for bias (-.3). Thus accuracy could not be confirmed. To assess precision, TempTraq and PA data were analyzed separately using repeated measures mixed models, which were regressed against patient variables. The variance of residuals, analyzed with an F-test, is a measure of precision. We found no significant difference (p = .2) between the two measures. Thus precision could be confirmed
Conclusions: Accuracy could be improved by adjusting for an off-set bias of -.30. Making this adjustment to the TempTraq and then doing a 2-arm trial (N =40) that allows comparison between the adjusted instrument with the original instrument could confirm this offset bias.
To read more about the TempTraq® thermometer Click Here.