Purpose:Ratings of perceived exertion (RPE) are widely used for monitoring individuals during graded exercise testing. Studies of the reliability of RPEs across various exercise conditions have produced mixed results. The purpose of this study was to assess the reliability of RPEs during graded exercise testing by comparing the perceptual-physiological relationship between the Bruce and Balke treadmill protocols throughout a broad range of relative exercise intensities.Methods:Thirty-eight middle-aged men and women completed two maximal treadmill graded exercise testing separated by 48 hours. Test order was randomly assigned. RPEs were compared across protocols and between gender at selected exercise intensities using a series of two-way analysis of variances with repeated measures.Results:A comparison of RPEs (Borg 15-point scale) during the graded exercise testing revealed significant protocol and gender differences at 40%, 60% and 80% of maximal heart rate reserve. RPEs were significantly higher during the Balke protocol compared to the Bruce at each intensity (40% = 9.5 ± 2.0 vs. 8.3 ± 1.6; 60% = 12.7 ± 2.4 vs. 11.1 ± 2.3; 80% = 15.7 ± 2.2 vs. 14.1 ± 2.0). In addition, men rated each intensity significantly higher than the women (P < 0.05).Conclusions:The results from the present study confirm that the perceptual-physiological relationship observed during graded exercise testing varies as a function of the treadmill protocol employed and that these differences extend throughout the exercise training intensity range (40-80% of maximal heart rate reserve) recommended for healthy adults. The perceptual differences between the protocols could not be accounted for by any of the physiological measures assessed within the study. These results have implications when using RPEs from exercise testing for exercise prescription purposes.