Introduction

Measuring body temperature is one of the oldest methods for identifying illness and evaluating efficacy of treatments; it remains a vital assessment of health today. Ideally, measurement of temperature should be minimally invasive, quick, reliable, accurate, safe, and minimize dependence on user technique. Accurate and reliable temperature assessment depends on physiologic and technical factors, user technique, and equipment maintenance.1

At the authors' multicenter community-based health system, the device of choice for noninvasive temperature assessment is an infrared-sensing temporal artery thermometer (TAT). To measure temperature, the user moves the thermometer across the skin overlying the temporal artery. The TAT repeatedly samples the skin temperature and runs an algorithm, providing a calculated temperature within seconds.1

Anecdotally, nurses at the organization were reluctant to rely on TAT results, especially when results were out of normal range or when fever was suspected. Based on their own experiences and observations of the practice of others, members of the Nursing Research Council (NRC) were concerned that there was inconsistency in user technique and clinicians often used another device to validate temperature readings. Patients may not have been treated appropriately-for example, a febrile patient overlooked-because incorrect TAT technique produced inaccurate temperature values.

Background

Previous research demonstrated conflicting findings on the reliability and validity of TAT temperature measurements. Some studies concluded that TAT measurements have acceptable agreement with standard measurements, including oral and esophageal.2-4 Others have concluded that a statistically and clinically significant variability exists between TAT measurements and various measurement modalities, including oral and tympanic.5-7

McConnell and colleagues are the only group to specifically evaluate the reliability of TAT measurements; they reported acceptable interrater and intrarater reliability.3 However, Penning and colleagues found greater discrepancy among TAT measurements when multiple individuals collected measurements compared with when a single researcher collected data.8 Additionally, Bahr and colleagues found that more than 70% of clinicians who received initial and supplemental education in TAT technique performed the measurements incorrectly. Only 50% of these clinicians were aware of proper maintenance techniques for the TATs.9

With the exception of Penning and colleagues, who allowed nurses deemed competent in the use of TATs to collect data for one arm of the study, all other references used expert raters to complete data collection.8 None of the studies explored the relationship between technique and validity or reliability of the results or observed actual clinical practice. As Davie and Amoore stated, "Technique and maintenance were significant factors in determining the accuracy and reliability of temperature measurements."1 It is possible that technique and maintenance were responsible for the discrepancy in findings of studies evaluating the reliability and validity of TAT measurements.

Objectives. The purpose of this study was for the NRC to observe the frequency of correct TAT technique among nurses who self-report TAT use and to evaluate differences in temperature measurements between correctly and incorrectly obtained TAT temperatures.

Methods

Design, setting, and sample. After receiving Institutional Review Board approval, NRC members conducted this observational study at an education session for nursing leadership across the system. This session was attended by approximately 150 nurses from staff and leadership; only those who reported recent experience (within the past 5 years) with the TAT were included as potential subjects. Utilizing mean differences in TAT measurements reported in previous studies of 0.5[degrees] F, and power of 0.8, the sample size required was 17 subjects.3 To achieve this, subjects participated in this study at multiple stations. Each station aimed to recruit a minimum of 17 subjects. However, post-hoc analysis revealed the standard deviation of the sample to be greater than 0.5, requiring a larger sample size for statistical analysis. Because no statistically significant differences in mean temperature or standard deviation among the stations were observed, the data were collapsed and analyzed as a whole.

Procedure. Four stations were set up to obtain TAT readings. Each station was obscured by a curtain to ensure privacy. An NRC member invited attendees to participate, shared the consent information sheet, and directed nurses who verbally consented to available stations. Two members of the NRC covered each station. One member of the NRC had her temperature taken by participants, while the other member evaluated the user's technique as either correct or incorrect.

Prior to the event, the evaluators were trained to assess TAT technique. A 10-point skill checklist was used to evaluate technique, and an error in any step of the process was categorized as incorrect. The NRC member documented the temperature in the appropriate column (correct technique versus incorrect technique) on the data collection tool.

Because the number one reason for low TAT readings is a dirty lens, investigators ensured that the lens of each TAT device was clean before data collection. In addition to observing TAT technique, observers asked each participant to state the technique and interval for cleaning the TAT lens. Observers recorded whether the participant was or was not aware of the correct technique on the data collection tool.

Results

A total of 69 subjects were enrolled in this observational study. Thirty-nine percent of the subjects (n = 27) demonstrated correct technique in all categories on the checklist. These participants obtained temperatures that ranged from 97.2[degrees] F to 98.9[degrees] F (36.2[degrees] C to 37.2[degrees] C) with a mean of 97.7[degrees] F (36.5[degrees] C). The remaining subjects (n = 42) failed to use correct technique in at least one category on the checklist. The most common errors included scanning the forehead only and skipping the temple, or scanning straight from the forehead to under the ear without lifting the thermometer as recommended. Temperature readings for this group ranged from 75[degrees] F to 100[degrees] F (23.9[degrees] C to 37.8[degrees] C) with a mean of 95.2[degrees] F (35.1[degrees] C). Because some of the temperatures obtained were beyond reasonable clinical limits, six outliers were removed. The resulting mean temperature for incorrect technique was 97.6[degrees] F (36.4[degrees] C).

The standard deviation among scores for nurses using correct technique was 0.534[degrees] F and 6.58[degrees] F for those who used incorrect technique. Using Levene's test to assess the equality of variances, the standard deviation of incorrect technique was significantly greater than correct technique (p < 0.01). This difference was also noted after removal of outliers (p = 0.036). The results suggest that variability in TAT results is primarily related to user technique. Using correct technique to obtain a TAT reading produces more credible results with less variability. Further, only 28% (n = 19) were aware of correct TAT cleaning procedures.

Discussion

Based on Penning and colleagues' research, the expected finding was that the TAT temperatures taken using the correct method should be an acceptable way to gauge temperature and is as accurate as a rectal temperature in healthy individuals.8 Bahr and colleagues' research, however, showed that many practitioners were not performing the measurement correctly (over 70%, according to one study, and 61% in this study), and only about 50% were aware of proper maintenance techniques for TATs.9 This led us to expect that many of our clinicians might be unaware of both the proper technique and the proper cleaning method.

To improve the accuracy of TAT temperature readings, clinicians must utilize proper technique. It is important that they understand the significance of improper technique and maintenance to reliable temperature assessment. When a questionable temperature reading is obtained, awareness of the importance of technique could encourage clinicians to reevaluate and reeducate themselves on proper TAT technique. Within the study organization, these findings have been disseminated, and strategies have been implemented to support correct technique at the point of care, including attaching a diagram of how to take a TAT measurement correctly to thermometers, and establishing a device cleaning schedule to ensure that cleaning requirements are met.

Conclusion

TAT measurements have many advantages in the clinical setting: They are minimally invasive, quick, and safe. Our results support the hypothesis that user technique significantly impacts the reliability and validity of TAT readings. The results of this study highlight the importance of correct technique and the imperative that nurses understand the impact of incorrect results. At times, it may be more appropriate to reevaluate user technique than to question the reliability of the device.

Additional studies are needed to further support these results, including comparing TAT with oral and rectal thermometer readings. Further research could explore barriers to proper TAT technique adherence at the bedside. In addition, the study could be replicated with a patient population with the potential for greater variability in temperature readings, such as patients known to have fever or hypothermia. Further, the findings of this study should be evaluated in clinical practice.

REFERENCES