Prediction of one-repetition maximum from submaximal ratings of perceived exertion in older adults pre- and post-training.

Background: Individual's one-repetition maximum (1-RM) is required to calculate and prescribe intensity for resistance training, while testing protocols enhance the risk of injuries and are time-consuming.

Aims: The aim of the present study was to assess the accuracy of 1-RM prediction from ratings of perceived exertion (RPE) of resistance exercises performed at submaximal sets (intensity and volume) in older adult males before and after a 12-week rehabilitation program.

Methods: 18 untrained subjects (70.4 ± 4.5 years) first completed a 1-RM direct assessment with a horizontal leg press pre- and post-training. Thereafter, participants performed, in a random order, 2-repetition sets with loads unknown to them (corresponding to 20, 45 and 70 % of 1-RM). The RPE was recorded immediately after the sets. That RPE associated to its corresponding load was subjected to a linear regression analysis to extrapolate the maximal RPE score and its corresponding 1-RM.

Results: RPE and relative intensities of sets appeared related pre- [r (2) = 0.59, standard error of estimate (SEE) = 13.3 %] and post-training (r (2) = 0.83, SEE = 8.1 %). Differences between measured and predicted 1-RM were reduced from the beginning to the end of training but standard deviations remained high (17.4 ± 11.8 vs. 4.2 ± 11.1 kg). Pre-training, 1-RM expressed relatively to body weight was negatively related with the errors of 1-RM predictions (r (2) = 0.39, p = 0.03).

Conclusions: In older subjects, RPE may be used to predict 1-RM; however, the predicted value deviates considerably from the measured one, necessitating cautious application. Importantly, this method allows to capture training-induced change in 1-RM, thus making possible assessing training's effectiveness and allowing its modification if necessary.