Measurement of heart rate during exercise is extremely important to the understanding of energy expenditure and physical training of the cardiovascular system. Accurate measurements of heart rate have been encumbered by the size of the monitor, difficulty of use, and the reliability of the measurement devices in the presence of movement artifacts. There is interest in developing accurate, small, non-intrusive monitors for use in exercise regimens where direct wire leads cannot be tolerated, where ease of use is critical, and where movements may give rise to artifacts. PURPOSE This study evaluated three telemetric heart rate monitors, the Actiheart [AH] (Mini Mitter Co., Inc., Bend, OR), Polar #S-810i and XR [PL](Polar, Finland), and XHR VitalSense Heart Rate sensor [VS] (Mini Mitter Co., Inc., Bend, OR). METHODS All heart Rate sensors were compared to the Welch-Allyn Atlas (6200 Series) stationary heart rate monitor. Six (6) volunteers, three men and three women, wore the four sensors, placed in specific locations on the chest. All sensors were attached to 3M Red Dot electrodes in a specific placement to maximize the electrical signal and placement was in a random order so as not to bias the data due to placement alone. Data were recorded as beats per minute in fifteen-second epochs. Subjects rested for ten minutes to determine resting heart rate and then all subjects completed a modified Bruce protocol treadmill test (3 minutes 1.7 mph, 0% slope; 3 minutes 2.5 MPH, 3% Slope; 3 minutes 3.4 MPH, 6%; and 3 minutes 4.2 mph, 9% slope). Following the exercise protocol subjects were allowed to rest until heart rate returned to resting values. RESULTS All three of the devices demonstrated a similar specificity, 0.97, 0.98, 0.97, for AH, PL, and VS respectively. However, PL demonstrated a significantly, p < 0.01, reduced sensitivity (0.85) as compared to AH (0.98) and VS (0.98). In addition, PL also showed a significantly greater number of false positive readings (15.2%) as compared to AH (2.34%) and VS (2.5%). The correlation between PL and the Atlas standard was significantly worse during exercise as compared to rest. CONCLUSION The results reported here indicate that there is a need for additional studies to validate accurate measurements of heart rate in exercising subjects, particularly when the exercise is rigorous. Supported by Mini Mitter, CO., Inc and U.S. Army DAMD17-01-C-0022