Reliability of temporal artery thermometers.
Review of the Literature
Literature was reviewed by searching primary health care-related databases (CINAHL, PubMed) from 1990 to 2013. Keywords used in the search related to temperature monitoring with a temporal artery thermometer (temporal artery thermometer, temporal artery temperature, temporal artery thermometer accuracy, TAT, thermometer reliability, forehead temperature, forehead thermometer, infra-red temperature measurement, and noninvasive temperature measurement).
Despite its increasing use in health care settings, well-designed studies on accuracy of this device are limited (Barringer et al., 2011; Calonder et al., 2010; Fountain, Goins, Scoles, Hartman, & Hays, 2008; Frommelt, Ott, & Hays, 2008; Langham et al., 2009; Lawson et al., 2007; Mangat, Standley, Provost, Vasconcelos, & White, 2010). The temporal artery device was more accurate than axillary (Langham et al., 2009; Lawson et al., 2007) or tympanic temperatures (Frommelt et al., 2008; Langham et al., 2009; Lawson et al., 2007) when compared to rectal or other core temperatures, but not as accurate as oral temperatures (Frommelt et al., 2008; Lawson et al., 2007). Of particular concern with prior studies, none have evaluated the intra-rater and inter-rater reliability of the device when used by hospital personnel. Evaluating reliability in the temporal artery thermometer is particularly important because it uses the same infrared technology for temperature measurement as the tympanic thermometer. The tympanic thermometer is no longer recommended for acute care clinical use (Bamason et al., 2012; Hooper & Andrews, 2006) because large temperature variations with the device are thought to be from users' inability to consistently place the probe in the ear canal (Erikson & Meyers, 1994; Giuliano, Scott, Elliot, & Giuliano, 1999; Klein et al., 1993). Only one study evaluated intra-rater reliability of the temporal artery thermometer in an outpatient setting (Mangat et al., 2010); however, little difference was found between sequential measurements by the same individual. No studies have evaluated inter-rater reliability (sequential measurements by two different individuals) in any setting, and none have evaluated intra- or inter-rater reliability in an acute care hospital setting.
Reliability is the reproducibility of the temperatures when measured by the same individual (intra-rater reliability) and by two or more different individuals (inter-rater reliability). Ideally, repeated measurements by the same or two different individuals would have little or no difference between temperatures. Because the temporal artery thermometer uses the same technology as the tympanic thermometer, a device which has been found to have poor reliability (Etikson & Meyers, 1994; Giuliano et al., 1999; Klein et al., 1993), reliability testing of the temporal artery device is needed prior to widespread use in hospitals.
The primary purpose of this study was to determine the intra-rater and inter-rater reliability of the temporal artery thermometer when used by health care providers on hospitalized adult patients. A secondary purpose was to determine the bias (mean temperature difference between two devices) and precision (standard deviation of the mean temperature difference) of the temporal artery thermometer compared to the reference standard oral electronic thermometer.
Materials and Methods
This study was conducted in a 400-bed not-for-profit community hospital in the Rocky Mountain region of the United States on a 37bed orthopedic and general surgical care unit. Study approval was obtained from the institution's investigational review board prior to data collection. Data collection was completed over 3 months.
A method-comparison design was used to compare temperatures obtained by two different operators. The primary outcome was the difference in the temporal artery temperature measurements between the repeated measurements (single user, two users). The secondary outcome was the difference in the temporal artery and oral electronic thermometer measurements.
Temporal artery temperature was obtained according to manufacturer's directions with a Temporal [R] Scanner, TAT 5000 (Exergen, Waterton, MA). Manufacturer's specifications include a clinical accuracy of [+ or -] 0.2[degrees] F ([+ or -] 0.1[degrees] C) for temperatures of 61[degrees]-110[degrees] F (16[degrees]-43[degrees] C). Hair on the forehead and neck was brushed aside and, with the temperature probe flush on the center of the forehead, the device activation button was depressed and held while the probe was moved straight across the forehead to the temporal area hairline. With the button still depressed, the probe was lifted from the forehead and touched on the skin behind the ear lobe in the soft conical depression on the neck just below the mastoid. The activation button then was released and the displayed temperature recorded.
Oral temperature was obtained according to manufacturer's directions with an electronic thermometer (SureTemp[R] Plus Model 690, WelchAllyn, San Diego, CA). Manufacturer's specifications include a clinical accuracy of [+ or -] 0.2[degrees] F ([+ or -] 0.1[degrees] C for temperatures of 80[degrees]-110[degrees] F (or 26.7[degrees]-43.3[degrees] C). The temperature probe was inserted gently into the right or left posterior sublingual pocket and held in that position until the temperature was displayed digitally on the device.
Subjects for this study were a convenience sample of adult surgical inpatients. Inclusion criteria included age greater than 18 and less than 85, cooperative, and able to follow directions for temperature measurement. Exclusion criteria included presence of forehead or neck wounds or dressings, cervical collars or halo traction, and facial tics or tremors; and smoking, eating, or drinking within the last 15 minutes. A minimum sample size of 34 subjects was determined a priori by power analysis for Student's t-test (Cohen, 1997; Mayr, Buchner, Erdfelder, & Faul, 2007). Effect size was 0.5 (moderate), alpha level of 0.05, and a power of 0.80.
Prior to data collection, temperature devices used in the study were calibrated according to manufacturers' directions and dedicated for use in the study. Two investigators were trained in the proper use of the devices based on manufacturers' directions. The two investigators were observed during use of each device to assure manufacturers' procedures for proper device use were followed carefully.
Demographic and patient characteristic data were obtained from the medical record in consenting subjects. During a time when therapeutic temperature measurement was required, two trained investigators measured all required study temperatures within a 10-minute period. The first investigator obtained two sequential temporal artery thermometer measurements with no more than 30 seconds between the two measurements. Immediately after completion of the temporal artery measurements, oral temperature also was measured. As soon as the first investigator left the participant's room, the second investigator entered the room and obtained two sequential temporal artery thermometer measurements with no more than 30 seconds between the two measurements. Immediately after completion of the temporal artery thermometer measurements, oral temperature was measured. Because prior studies suggested diaphoresis could affect device accuracy, the presence or absence of diaphoresis in the forehead area was documented by both investigators prior to temporal artery thermometer use.
The second investigator was blinded from the results of the first investigator's measurements by sequestering the second investigator in a separate room during temperature measurement by the first investigator. The results of the first investigator's temperature measurements were sealed in an envelope prior to exiting the participant's room and not revealed until after the recording of temperatures by the second investigator.
Data were summarized using descriptive statistics. Bias (mean differences between two temperature measurements) and precision (standard deviation [SD] for the mean difference scores) values for intra-rater and inter-rater reliability were calculated using standard formulas for mean and SD calculations. Mean difference and limits of agreement ([+ or -] 2 SD) between the two measurements with the temporal artery device obtained by the same investigators (intra-rater) and the two different investigators (inter-rater) were calculated and graphed according to the Bland-Altman method (Bland & Altman, 1986; Chatburn, 1996; Hanneman, 2008; Szaflarski & Slaughter, 1996). For inter-rater reliability, the first temporal artery temperature recorded by each investigator was used for calculations. To determine the reliability of the temporal artery thermometer compared to the oral thermometer, the first temporal artery temperature obtained by the first investigator was used for calculations. Acceptable reliability levels were set a priori at a SD of less than 0.6[degrees] F. This determination was based on experts' opinion in prior temperature studies evaluating the level of agreement between different temperature devices for clinical use (Bridges & Thomas, 2009; Giuliano et al., 1999; Lawson et al., 2007; Sessler, 2008). Acceptable levels for agreement between the temporal artery and oral device also were set a priori based on experts' opinion at a bias of less than or equal to 1.0[degrees] F and precision of less than or equal to 0.6[degrees] F (Bridges & Thomas, 2009; Giuliano et al., 1999; Lawson et al., 2007; Sessler, 2008). Regression analysis was used to determine if the presence of diaphoresis explained differences between temporal artery and oral temperatures. The level of significance was set at p < 0.05.
Temperatures for 34 surgical patients were measured once during their postoperative stay on a general surgical unit (see Table 1 for demographic and patient characteristics). Temperatures ranged from 97.3[degrees]-100.6[degrees]F during data collection, with five participants noted by one or both data collectors to have forehead diaphoresis at the time of temporal artery temperature measurement.
Intra- and Inter-Rater Reliability
Differences between the first and second temporal artery temperatures measured by the same investigator (intra-rater reliability) were 0.4[degrees]-1.3[degrees] F for the first investigator and 1.0[degrees]-1.0[degrees] F for the second investigator (see Figures 2A & 2B). Average differences were 0.14[degrees] ([+ or -] 0.43[degrees]) and 0.13[degrees] ([+ or -] 0.41[degrees]) F for the two investigators, respectively (see Table 2). Differences between the first temporal artery temperature measured by each investigator (inter-rater reliability) ranged from 1.10[degrees]-0.80[degrees]F, averaging 0.19[degrees] ([+ or -] 0.48) F (see Figure 2C). Intra-and inter-rater reliability levels were within the range of acceptable values set a priori (SD<1.0[degrees] F).
Agreement of Temporal Artery and Oral Temperatures
The difference between the temporal artery and oral temperatures was 0.48[degrees] ([+ or -]0.88[degrees]) F for the first investigator and 0.47[degrees] ([+ or -]0.57[degrees]) F for the second investigator. These values are within the range set a priori for acceptable levels of agreement between a reference thermometer (oral electronic) and a new temperature device (temporal artery). Diaphoresis was not found to explain differences between the temporal and oral temperatures (p>0.05).
In a reliability study of temporal artery thermometers in hospitalized patients, intra-reliability and interreliability values were < 0.5[degrees] F, within the range of values considered acceptable when evaluating temperature devices for hospital use. The differences in two temperatures measured by each investigator and between two investigators were consistent across the range of temperatures observed in this study, with only two temperature differences greater than 1.0[degrees] F in 102 pairs of temperature comparisons. As a secondary finding of this study, the level of agreement between the temporal artery and oral thermometer was within the range considered acceptable by experts in temperature measurement research. This is the first study to evaluate both intrarater and inter-rater reliability of the temporal artery thermometer in a hospital setting. Intra-rater reliability of the temporal artery temperature device was evaluated in one prior study, finding small differences between two measurements by the same individual (Mangat et al., 2010), similar to this study.
Prior to using a new temperature device in patient care, nurses must determine not only how well the thermometer agrees with temperatures measured with a reference thermometer, but also if measurements are similar when obtained sequentially by the same person (intra-rater reliability) and by more than one person (inter-rater reliability). Large differences between these measurements may indicate the device is difficult to use properly each time temperatures are obtained, introducing error into the temperature measurement. If intra-rater or inter-rater reliability is poor, use of the device is not recommended for clinical practice. This testing is particularly important with the temporal artery thermometer because reliability was poor in the tympanic thermometer, which uses a similar technology to measure temperature. The reliability data for this study indicate the temporal artery thermometer consistently provides similar temperatures when measured repeatedly by one operator and when measured by two different operators.
In addition to determining the reliability of the temporal artery thermometer, authors also studied the agreement of the device with oral electronic temperature measurements. Acceptable bias and precision values found in this study are similar to a number of well-controlled prior studies which compared the temporal artery thermometer to core temperature measures (pulmonary artery or bladder catheter thermistors) (Calonder et al., 2010; Langham et al., 2009; Lawson et al., 2007; Lee et al., 2011; Siberry, Diener-West, Schappell, & Karron, 2002) or oral electronic thermometers (Barringer et al., 2011; Fountain et al., 2008; Frommelt et al., 2008). Several other studies have not found the bias and precision of temporal artery thermometers to be within the acceptable range recommended by experts for clinical use (Kimberger, Cohen, Illievich, & Lenhardt, 2007; Marable, Shaffer, Dizon, & Opalek, 2009; Ronneberg, Roberts, McBean, & Center, 2008; Stelfox et al., 2010; Suleman, Doufas, Akca, Ducharme, & Sessler, 2002). However, each of these studies had one or more serious methodologic problems (poorly described methods, inappropriate reference thermometers, lack of device calibration prior to the study, incomplete technique when using the temporal artery thermometer, and/or inappropriate statistical analyses) that limited the generalizability of those study findings. Reviews of the literature and/or meta-analyses do not include these studies because they failed to meet the methodologic inclusion criteria (Bamason et al., 2012; Hooper & Andrews, 2006).
Several limitations were present in this study. Only two different investigators were used to determine inter-rater reliability. Greater imprecision of the device may be found with multiple users, a common situation in most acute care practice situations. Another limitation was the lack of abnormal body temperatures at the time of device testing. Finally, the procedure for using the temporal artery temperature was based on the manufacturer's direction for use, including moving the device to behind the ear prior to completion. Not complying with all the steps for measurement may lead to greater errors in temperature measurement than observed in this study. Given the busy nature of most acute care units today, coupled with unlicensed personnel doing the majority of temperature measurements, inadequate technique could occur.
Implications for Practice
Based on the results of this study, temporal artery thermometers appear to be a reliable way to measure temperature noninvasively in hospitalized patients. While oral electronic temperatures are used commonly in acute care settings, the temporal artery thermometer provides an additional accurate method for temperature assessment. This device could be used in clinical situations where oral temperature assessment is contraindicated for safety reasons (e.g., alterations in level of consciousness, combative behavior) or impediments to device use (e.g., oral trauma or surgery, mucositis), and would be a more accurate method than obtaining axillary temperatures in these situations.
As is true with any medical equipment, care should be taken to assure personnel using the device follow correct procedures for temperature measurement. Proper use of this device requires movement of the temperature probe across the forehead, as well as placement of the probe behind the ear, while still depressing the activation button. For accurate readings to be obtained, clinicians must include the ear probe touch. Nurses educating staff about the proper use of this device must emphasize this aspect of proper device use.
In this study, intra-rater and inter-rater reliability of the temporal artery thermometer was within range of values recommended by experts for use in clinical care. These findings indicate there is little user error associated with use of the temporal artery thermometer. In addition, the study found an acceptable level of agreement for temperatures obtained with the temporal artery thermometer compared to the reference thermometer (oral electronic).
Barnason, S., Williams, J., Proehl, J., Brim, C., Crowley, M., Leviner, S., ... Storar, A. (2012). Emergency nursing resource: Non-invasive temperature measurement in the emergency department. Journal of Emergency Nursing, 36, 523-530.
Barringer, L., Evans, C., Ingrain, L., Tisdale, P., Watson, S., & Janken, J. (2011). Agreement between temporal artery, oral, and axillary temperature measurements in the perioperative period. Journal of Perianesthesia Nursing, 25(3), 143-150.
Bland, J., & Altman, D. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 1, 307-310.
Bridges, E., & Thomas, K. (2009). Noninvasive measurement of body temperatue in critically ill patients. Critical Care Nurse, 29(3), 94-97.
Calonder, E.M., Sendelbach, S., Hodges, J.S., Gustafson, C., Machemer, C., Johnson, D., & Reiland, L. (2010). Temperature measurement in patients undergoing colorectal surgery and gynecology surgery: A comparison of esophageal core, temporal artery, and oral methods. Journal of Perianesthesia Nursing, 25(2), 71-78.
Chatburn, R. (1996). Evaluation of instrument error and method of agreement. Journal of American Association of Nurse Anesthetists, 64(3),261-268.
Cohen, J. (1997). Statistical power analysis for the behavioral sciences (revised ed.). New York, NY: Academic Press.
Erickson, R., & Meyers, L. (1994). Accuracy of infrared ear thermometry and other temperature methods in adults. American Journal of Critical Care, 3, 40-54.
Fountain, C., Goins, L., Scoles, D., Hartman, M, & Hays, V. (2008). Evaluating the accuracy of four temperature instruments on an adult inpatient oncology unit. Clinical Journal of Oncology Nursing, 12(6), 983-987.
Frommelt, T., Ott, C., & Hays, V. (2008). Accuracy of different devices to measure temperature. MEDSURG Nursing, 17(3), 171-174, 176, 182.
Giuliano, K., Scott, S., Elliot, S., & Giuliano, A. (1999). Temperature measurement in critically ill oral intubated adults: A comparison of pulmonary artery core, tympanic, and oral methods. Critical Care Medicine, 27, 2188-2193.
Hanneman, S.K. (2008). Design, analysis and interpretation of method-comparison studies. AACN Advanced Critical Care, 19, 223-234.
Hooper, V.D., & Andrews, J.O. (2006). Accuracy of noninvasive core temperature measurement in acutely ill adults: the state of the science. Biologic Research for Nursing, 8(1), 24-34.
Kimberger, O., Cohen, D., Illievich, U., & Lenhardt, R. (2007). Temporal artery versus bladder thermometry during perioperative and intensive care unit monitoring. Anesthesia and Analgesia, 105, 1042-1047.
Kistemaker, J., Hartog, D., & Daanen, H. (2006). Reliability of an infrared forehead skin thermometer for core temperature. Journal of Medical Endoscopic Technology, 30, 252-261.
Klein, D., Mitchell, C., Petrines, A., Monroe, M.K., Oblak, M., & Ross, B. (1993). A comparison of pulmonary artery, rectal, and tympanic membrane temperature measurements in the ICU. Heart and Lung, 22, 435-441.
Langham, G.E., Maheshwari, A., Contrera, K., You, U., Mascha, E., & Sessler, D.I. (2009). Noninvasive temperature monitoring in postaneshtesia care units. Anesthesiology, 111, 90-96.
Lawson, L., Bridges, E., Ballou, I., Eraker, R., Greco, S., Shively, J., & Sochulak, V. (2007). Accuracy and precision of noninvasive temperature measurement in adult intensive care patients. American Journal Critical Care, 16, 485-496.
Lee, G., Flannery-Bergey, D., Randall-Rollins, K., Curry, D., Rowe, S., Teague, M., ... Schroeder, S. (2011). Accuracy of temporal artery thermometry in neonatal intensive care infants. Advances in Neonatal Care, 11(1), 62-70.
Mangat, J., Standley, T., Provost, A., Vasconcelos, J., & White, P. (2010). A comparison of technologies used for estimation of body temperature. Physiologic Monitoring, 31, 1105-1118.
Marable, K., Shaffer, L.E., Dizon, V., & Opalek, J.M. (2009). Temporal artery scanning falls short as a secondary, noninvasive thermometry method for trauma patients. Journal of Trauma Nursing, 16(1), 41-47.
Mayr, S., Buchner, A., Erdfelder, E., & Faul, F. (2007). A short tutorial of GPower. Tutorials in Quantitative Methods for Psychology, 3(2), 51-59.
Ronneberg, K., Roberts, W.O., McBean, A.D., & Center, B.A. (2008). Temporal artery temperature measurements do not detect hyperthermic marathon runners. Medical Science for Sports Exercise, 40, 1373-1375.
Sandlin, D. (2003). New product review: Temporal artery thermometry. Journal of Perianesthesia Nursing, 18, 419-421.
Sessler, D. (2008). Temperature monitoring and perioperative thermoregulation. Anesthesiology, 109(2), 318-338.
Siberry, G., Diener-West, M., Schappell, E., & Karron, R. (2002). Comparison of temple temperature with rectal temperatures in children under two years of age. Clinical Pediatrics, 41(6), 405-414.
Stelfox, H., Straus, S., Ghali, W., Conly, J., Laupland, K., & Lewin, A. (2010). Temporal artery versus bladder thermometry during adult medical surgical intensive care monitoring: An observational study. BMC Anesthesiology, 10, 13.
Suleman, M., Doufas, A., Akca, O., Ducharme, M., & Sessler, D. (2002). Insufficiency in a new temporal-artery thermometer for adult and pediatric patients. Anesthesia and Analgesia, 95(1), 67-71.
Szaflarski, N., & Slaughter, R. (1996). Technology assessment in critical care: Understanding statistical analyses used to assess agreement between methods of clinical measurement. American Journal of Critical Care, 5, 207-216.
Erica McConnell, Deborah Senseney, Sheryl S. George, and Debra Whipple Erica McConnell, BSN, RN-ONC, is Staff Nurse, Orthopedic and Surgical Unit, Exempla Lutheran Medical Center, Wheat Ridge, CO.
Deborah Senseney, MSN, RN-BC, is Medical-Surgical Clinical Specialist, Exempla Lutheran Medical Center, Wheat Ridge, CO.
Sheryl S. George, MSN, RN-BC, is Orthopedic Clinical Nurse Specialist, Exempla St. Joseph's Medical Center, Denver, CO.
Debra Whipple, RN, is Former Staff Nurse, Orthopedic and Surgical Unit, Exempla Lutheran Medical Center, Wheat Ridge, CO.
TABLE 1. Demographic and Patient Characteristics Study Sample Age (average [+ or -] SD years) 58.9 [+ or -] 17.2 Sex Male 18 Female 16 Surgical Procedure Orthopedic 8 Other 26 Presence of Forehead No 29 Diaphoresis During Yes 5 Temperature Measurement TABLE 2. Average Temperature Differences ([+ or -] SD) and Number of Temperature Differences > [+ or -]1.0[degrees] F Between Two Sequentially Measured Temporal Artery Temperatures (TAT) by the Same Investigator (Intra-Rater Reliability) and Two Different Investigators (Inter-Rater Reliability) in 34 Surgical Patients Number Temperature Mean [+ or -] SD Differences Temperature >[+ or -]1.0 (degrees F) [degrees] F Difference of 1st and 2nd TAT 0.14 [+ or -]0.43 0 for Investigator #1 (intra-rater reliability) Difference of 1st and 2nd TAT 0.13 [+ or -]0.41 1 for Investigator #2 (intra-rater reliability) Difference of 1st TAT of -0.19 [+ or -]0.48 1 Investigators #1 and #2 (inter-rater reliability)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research for Practice|
|Author:||McConnell, Erica; Senseney, Deborah; George, Sheryl S.; Whipple, Debra|
|Date:||Nov 1, 2013|
|Previous Article:||Developing a suicide precaution procedure.|
|Next Article:||Circulation savers: thrombolytic therapy.|