CONTEXT: The protective equipment worn during American football has been shown to increase thermal strain; however, the perception of this increased heat has not been examined. OBJECTIVE: To evaluate perceptual responses of American football players while wearing different uniforms during exercise in the heat and to evaluate how these responses may be used to monitor athlete safety. Design: Randomized controlled trial. Setting: Human Performance Laboratory. PATIENTS OR OTHER PARTICIPANTS: Ten men with more than 3 years of competitive experience as football linemen (age = 23.8 +/- 1.3 years, height = 183.9 +/- 1.8 cm, mass = 117.4 +/- 3.5 kg, body fat = 30.1% +/- 1.7%) participated. Intervention(s): On 3 occasions in hot, humid (33 degrees C, 48%-49% relative humidity) environmental conditions, participants completed 10 minutes of strenuous repetitive box lifting (RBL), 10 minutes of seated rest, and up to 60 minutes of treadmill walking. At each trial, they wore a different uniform condition: control (CON) clothing comprising shorts, socks, and sneakers; partial (PART) National Football League (NFL) uniform comprising the uniform without helmet or shoulder pads; or full (FULL) NFL uniform. Exercise, meals, and hydration status were controlled. MAIN OUTCOME MEASURE(S): Rectal temperature (T(re)), skin temperature (T(sk)), rating of perceived exertion (RPE), thermal perception (THM), perception of thirst (TST), and perception of muscle pain (MPN) were obtained for time points matched across trials. RESULTS: Nineteen of the 30 trials ended before 60 minutes of treadmill walking as a result of participant exhaustion. Mean treadmill time was longer for the CON condition (51.7 +/- 13.4 minutes) than for the PART (43.1 +/- 15.6 minutes; t(9) = 3.092, P = .01) or the FULL (36.2 +/- 13.2 minutes; t(9) = 4.393, P = .002) conditions. Neck and forearm T(sk) increased between the initial time point and the end of exercise in the PART (33.6 +/- 0.9 degrees C and 35.0 +/- 0.6 degrees C, respectively; F(2,18) = 9.034, P < .001) and the FULL (33.4 +/- 0.9 degrees C and 35.2 +/- 0.6 degrees C, respectively; F(2,18) = 21.011, P = .002) conditions. Rate of T(re) rise was greater in the FULL (0.042 +/- 0.010 degrees C/min) than in the PART (0.034 +/- 0.006 degrees C/min) condition (F(2,27) = 10.69, P = .04). We found a relationship at the post-RBL and final time points between RPE and THM (r = 0.75, P < .001 and r = 0.59, P < .001, respectively), RPE and TST (r = 0.76, P < .001 and r = 0.61, P < .001, respectively), and RPE and MPN (r = 0.63, P < .001 and r = 0.64, P < .001, respectively). The RPE was greater at the end of exercise in the PART (17 +/- 2) and FULL (18 +/- 1) conditions than in the CON (15 +/- 3) condition (F(2,18) = 7.403, P = .005).