One universally accepted component of an effective strength training program or exercise prescription is the overload principle. In essence, an individual must be exposed to increasingly more rigorous metabolic, physiologic, and psychological stimuli to manifest desired adaptations (12). However, a primary though often neglected component necessary for successful training is the level of recovery attained before initiating subsequent bouts of training (1,13). Despite the potential value and importance of monitoring an athlete's recovery status, there are few options that are adequate or convenient for monitoring day-to-day recovery (1). Further, there exist few (if any) valid and expeditious assessment instruments designed to evaluate an individual's level of recovery before initiating subsequent bouts of physical training (1,23). This is problematic in that insufficient recovery can precipitate detrimental side effects leading to suboptimal performance and, ultimately, chronic overtraining (11,12,14,16).
Although identifying the amount of time needed to ensure optimal recovery has received some attention (4,10,13,15,22), studies have used lengthy and exhaustive measures potentially viewed as impractical for utility in daily training. There exist laboratory-based physiological measures (e.g. O2max, running economy, lactate threshold, etc.) used for identifying overtraining; however, these tests are also impractical for monitoring day-to-day recovery status. Limited attention has been given to the derivation of valid field-based tests designed to quickly and accurately assess an individual's recovery status. Coutts et al. (2) have identified a number of field-based tests designed to assess an individual's level of recovery such as the maximal 3-km time trial run, submaximal heart rate (HR) test, and 5-bound tests. However, these particular tests are intended to identify markers of overtraining and not, necessarily, to assess the day-to-day recovery status of the athlete because they are physically taxing and time consuming to perform. Ramifications of tests involving high levels of exertion on an athlete's training program include gross disruption of effective training on the day the such tests are administered to determine recovery status.
In 1998, Kentta and Hassmen (11) attempted to create a practical, noninvasive method of monitoring recovery status by creating the total quality recovery (TQR) scale. A detailed review of the procedures associated with the TQR is presented in Kentta and Hassmen (11). The TQR is analogous to Borg's 6-20 ratings of perceived exertion (RPEs) scale and is comprised of 2 components: perceived recovery and action recovery. The perceived component requires the athlete to record their perceived level of recovery immediately before going to sleep at night, whereas the action component is more detailed and consists of the athlete awarding themselves points in various categories such as nutrition and hydration, sleep and rest, relaxation and emotional support, and stretching and active rest (11,12). Ultimately, points accrued from each respective scale (i.e., perceived and action) are subjected to calculations (outlined in an accompanying TQR manual) to determine their respective level of recovery. Although innovative, this system is untested in the published literature and more importantly, tedious to a level which would likely elicit low compliance (and therefore questionable application) among athletes. That notwithstanding, Kentta and Hassmen (11) have documented that the potential for monitoring a subjective estimation of recovery between training sessions may be as applicable as monitoring sense of effort using RPE during training bouts. To that end, the scientific rationale guiding the TQR scale is novel, but perhaps the exacting and meticulous procedures (i.e., point assignment and calculation of scores) associated with its use mitigates the convenience typically associated with subjective measures. Lack of convenience apparently deterred its adoption by the sport and exercise community over the last 10 years.
Recovery is most assuredly an integration of physiological, psychological and emotional responses (2,11,18). Development of a tool that concurrently considers these important components is highly attractive. Rating of perceived exertion has consistently been shown to be a useful tool for gauging individual exercise tolerance and has been linked to a myriad of physiological mediators such as: ventilation, oxygen uptake, blood lactate concentration, and HR (5-8,17,21), rather than being linked to any single mediator. Because RPE has demonstrated exceptional utility for assessing psychobiological status during exercise (5-8), it is plausible that a parallel measurement tool used to assess recovery status would prove to be a functional and practical instrument. Further, a perceptually based tool can be used daily to noninvasively monitor an athlete's recovery status without disrupting that day's training schedule. With perceptual measures representing a gestalt attributable to a myriad of factors, a potential advantage of this paradigm for assessing recovery is that the likelihood of detecting overtraining is considerably greater (vs. reliance on a single indicator such as resting HR). Therefore, the purpose of this study was to create and assess the efficacy of a perceived recovery status (PRS) scale to assess the level of recovery after exhaustive exercise after varying recovery durations. The primary goal of the study was to examine the utility of a PRS scale in allowing an individual to subjectively estimate their level of recovery relative to a subsequent exercise performance. It was hypothesized that an individual's self-selected level of perceived recovery using a newly developed PRS scale would share a significant correlation with subsequent exercise performance.
Experimental Approach to the Problem
To create a simple and valid field assessment of an individual's recovery status, a PRS scale was created (Figure 1). The PRS scale is a 0-10, scalar representation of varying levels of an individual's level of PRS, similar to that of an RPE scale. We chose this form of representation because RPE is a commonly used and easily understood measure of an individual's perception of effort during exercise; thus, the perceived feeling of recovery should transfer well using a similar scale. The subjects were read and given standardized instructions explaining how to interpret the PRS scale and the numerical and verbal anchors contained within it. Also, subjects were presented a continuous 100-mm visual analog scale (VAS) with numerical markers ranging from 0 to 10 marked at each 10-mm interval. Verbal descriptors were presented of “not at all recovered” at 0, “moderately recovered” at the 50-mm mark, and “very well recovered” at the 100-mm mark. The subjects were asked to draw a vertical line that intersected the horizontal descriptor scale at the appropriate position that best describes their perceived level of recovery. Thereafter, each subject completed exercise sessions consisting of 3 cycles of 8, 30-m maximal sprints after either 24, 48, or 72 hours of recovery.
Subjects were always asked to estimate their perceived level of recovery before and after a standardized warm-up, before performing a subsequent bout of intermittent work. The level of perceived recovery was then compared with the change in performance during each bout of exercise relative to their previous trial. The exercise protocol with the varying recovery periods (i.e., 24, 48, and 72 hours) was designed to expose subjects to exercise sessions in which they were not fully recovered or able to optimally perform 24 hours after the baseline trial but also afford them enough recovery time (i.e., 48 or 72 hours) to achieve a higher recovery status and perform optimally in subsequent trials (10,15). Thus, each individual was asked to perform an identical exercise session under conditions in which only the recovery duration from the previous bout was manipulated. Ultimately, it was hypothesized that this would allow for greater inference regarding the utility of the PRS under conditions when subjects were perhaps “underrecovered” and when they were progressively more recovered.
Sixteen (8 men, 8 women) subjects provided written, informed consent before testing in accordance with the local institutional review board (IRB). All individuals were at least moderately active (assessed via questionnaire) and participated in intermittent high-intensity work at least once a week. An individual was considered moderately active if they participated in at least 30 minutes of exercise at least 4 days per week, whereas intermittent high-intensity work was defined as any activity that incorporated undulating periods of maximal, or near maximal, work integrated with episodes of low-to-moderate exercise. Before data collection, all individuals were assessed for height (cm) and total body mass (kg) using a calibrated stadiometer and beam scale, with body fat percentage estimated using the 3-site method (men: chest, abdomen, and thigh; women: tricep, iliac, and thigh) (20) by skinfold calipers (Lange, Cambridge, MD) (Table 1). All subjects arrived at the university recreational fields at least 3 hours postabsorptive and were instructed to be adequately hydrated, to abstain from caffeine for at least 4 hours, and alcohol 24 hours, before each testing session. Also, subjects were instructed to eliminate any structured exercise bouts (except for the experimental trials) throughout the data collection period, which lasted a total of 6 days. Subjects were also asked to get adequate sleep and to replicate dietary intake on days before exercise testing. Before beginning each trial, subjects were queried regarding the adherence to the guidelines about dietary issues, sleep, and physical activity. Also, each individual was asked to report any previously existing illness, injury, or any other physical or emotional issue that would hinder their performance. Criteria for exclusion from the study involved the acknowledgment or observed evidence of any medical or orthopedic problem severe enough to disrupt the subject's performance or endanger his or her health or a self-reported fitness classification below moderately active.
To determine the utility of a PRS scale as a convenient, noninvasive marker of recovery relative to subsequent training, each subject completed 4 separate trials of repeated maximal sprinting on 4 separate days using different recovery periods each trial. Each subject completed a familiarization session that consisted of 3 cycles of 8 repetitions of 30-m sprints (Figure 2) at least 1 week before beginning experimental trials. No perceptual, performance, or physiological data were collected. Each subject's foot position, as an individual preference, was recorded during the familiarization session and replicated throughout the trials.
After familiarization, subjects completed a baseline and 3 subsequent trials of repeated maximal sprint exercise. All testing took place outdoors on a level grass playing surface during fall months in the southeastern United States. Before each exercise trial, the environmental conditions were assessed and there were no significant differences between trials for air temperature, humidity, or wind speed (data not shown). The intermittent exercise protocol consisted of 3 cycles of 8 (total of 24 sprints) 30-m maximal-effort sprints and was conducted on 4 separate days with each trial being separated by 24, 48, or 72 hours after each previous bout. The recovery periods were assigned in a counterbalanced order after completion of the baseline trial. This protocol of assessing recovery after fatiguing exercise has been used in previous studies investigating recovery and exercise performance (10,15).
The protocol of the repeated 30-m intermittent sprints is detailed in Figure 2. On each testing day, subjects arrived at the same time of day and performed a warm-up that consisted of ∼300 m of light running, 5 minutes of dynamic stretching exercises (e.g., high-knee, carioca) followed by 2 practice repetitions of the 30-m intermittent sprint. After the warm-up, subjects performed 3 cycles of intermittent sprinting. Each cycle consisted of 8 electronically timed (Speed trap II wireless timing system, Power-Systems Inc., Knoxville, TN) 30-m maximal sprints followed by a 45-second recovery period that incorporated a 10-m “easy” jog (i.e., deceleration) followed by a 10-m walk with the remainder of the recovery period being passive (Figure 2). After each 45-second rest period, subjects were prompted to immediately perform another sprint. Subjects were instructed to approach the line 10 seconds before initiating each sprint and were given a 5-second countdown to signal the start. The infrared timing system was set at an appropriate lower-leg height, as per the manufacturer's instructions, and was used to record the duration to the nearest hundredth of a second of each 30-m maximal-effort sprint. Additionally, each segment of the sprint protocol was marked by cones set at the appropriate distance.
Subjects were given a 5-minute rest period after completing each set of 8 sprints. Before initiating each sprint, RPEs using the omnibus (OMNI) scale were recorded. Heart rate was recorded before and immediately after each sprint using an HR monitor (Polar Inc, Port Washington, NY). After the completion of each cycle of 8 sprints, each individual's blood lactate concentration was assessed using capillary blood samples taken from the fingertip and analyzed using an enzymatic portable lactate system (Lactate Pro, Arkray Inc, Kyoto, Japan). The system was calibrated in accordance with manufacturer's instructions before each trial, and each blood draw was analyzed in duplicate to ensure reliability (±1 mmol). Also, subjects were asked to provide a session RPE (S-RPE) using the OMNI scale to rate the global difficulty of the exercise bout (3) approximately 20 minutes after each trial.
Each recovery trial yielded 3 markers of recovery. The first marker of recovery was an individual's level of perceived recovery obtained from the PRS value before beginning each trial. The second marker was the perceived level of recovery as assessed by the 100-mm VAS, whereas the third marker of recovery was a computed variable of change in sprint time. After each recovery trial, individuals having achieved an identical, faster, or slower total time sprinting relative to their previous performance was calculated to evaluate the level of recovery. Specifically, the second trial (a recovery trial) was compared with the first (baseline trial), whereas the third trial was compared with the second and the fourth trial to the third. Thus, there would be an expected change in sprint time for each subsequent recovery bout.
To determine the relationship between the PRS scale and the 100-mm recovery scale (both pre- and post-warm-up) and their respective utility for assessing recovery relative to subsequent performance was analyzed using a Pearson product moment correlation coefficient (r). That is, a correlation was performed to identify the relationship between each individual's PRS scale value given before each recovery trial and the subsequent change in sprint time during the recovery trials. The a priori level of practical significance of the r was set at −0.70 or better. To evaluate any significant difference between the expected and observed outcomes of PRS score and change in performance, a chi-square analysis was also performed.
Rating of perceived exertion, HR, and [La] response were analyzed using a series of 4 (trial) × 3 (cycle of 8 sprints) repeated-measures analysis of variance (ANOVA). Session RPE and total sprint time for each trial were analyzed using a 1-way repeated measure ANOVA. When appropriate, Fisher's least significant differences (LSD) post hoc analyses were performed to identify any significant differences. The statistical power (N - B) and effect sizes (η2) were also calculated and are reported. All data were analyzed using SPSS (v. 16.0) statistical platform and reported as the mean ± SD. Statistical significance was determined a priori at the 0.05 level of significance.
The PRS value recorded post warm-up demonstrated the strongest (albeit moderate), significant (p < 0.01) correlation (r = −0.63) between PRS and change in performance (Figure 3). This correlation coefficient failed to meet our a priori level of practical significance (r = −0.70). The correlation analyses revealed that the PRS scale measurement taken pre-warm-up produced a weak-to-moderate correlation coefficient of r = −0.41 between recovery status and sprint performance. An intraclass correlation analysis between pre- and post-warm-up PRS scores revealed an r value of 0.37, suggesting only weak-to-moderate agreement between the 2 measurements. Additionally, correlation analyses revealed that change in sprint time and the VAS (both pre- and post-warm-up) yielded considerably weaker correlation coefficients of r = −0.13 and −0.02, respectively, when compared with the change in sprint performance and the PRS scale given either before or after a warm-up. Because of not only the strength of the PRS measurement taken post-warm-up, but also the weak relationship between change in sprint performance and PRS when using other scales, only results concerning the PRS taken post warm-up are reported.
Individual data and the overall accuracy of the PRS scale to identify exercise sessions yielding improved or declined sprint performance is shown in Tables 2 and 3. Using the presented data, both sensitivity and specificity of the PRS scale were calculated. Sensitivity refers to the percent of individuals reporting an expectancy of declined sprint performance before a trial relative to the total number of individuals demonstrating declined sprint performance, whereas specificity is the percentage of individuals reporting PRS scores indicating an expectancy to produce improved sprint performance relative to the number of people producing improved sprint performance. The resulting sensitivity and specificity scores derived were 82 and 81%, respectively. Results from the chi-square analysis revealed that there was a significant difference between the expected (because of chance) and observed outcomes between change in sprint performance relative to level of perceived recovery (χ2 = 15.7; p < 0.01).
The average, SD, range and 95% confidence limits associated with the PRS is shown in Table 4. In general, a PRS greater than 5 seemed to yield (largely) an improved performance, whereas a PRS equal to 5 tended to yield either improved or diminished performance. A PRS value lower than 5 generally predicted that an individual did not perform as well as the previous bout of exercise.
The results from the repeated-measures ANOVA revealed no significant main effect between trials for sprint time (p = 0.61). A significant main effect was found between cycles of sprints for sprint time (p < 0.01; η2 = 0.62; N − B = 1.0). There was a significant difference (p < 0.05) in total sprint time observed between cycles, with cycle 3 producing significantly slower sprint times (42.0 ± 0.13 seconds) than both cycle 2 (41.7 ± 0.10 seconds) and cycle 1 (41.2 ± 0.27 seconds), with cycle 2 significantly slower than cycle 1.
Results from a series of repeated-measures ANOVA revealed no significant main effect among trials for RPE (p = 0.71) but a significant main effect among the 3 cycles of 8 sprints (p < 0.01; η2 = 0.79; N − B = 1.0). Univariate post hoc analyses showed that cycle 1 RPEs (2.8 ± 1.2) were significantly lower (p < 0.01) than both cycle 2 (4.6 ± 1.2) and cycle 3 (5.9 ± 1.2) for all 4 trials. Also, RPE values recorded during cycle 2 were significantly lower (p < 0.01) than RPEs during cycle 3 for all trials.
Similarly, there was no significant main effect among trials for HR (p = 0.66) but a significant main effect among cycles (p = 0.01; η2 = 0.45; N − B = 0.95). Post hoc tests revealed that, similar to RPE, cycle one HR (169 ± 10 b·min−1) was significantly lower (p < 0.01) than cycle 2 (172 ± 9 b·min−1) and cycle 3 (174 ± 10 b·min−1). Subsequently, cycle 2 HR values were significantly lower (p = 0.01) than the HR values observed during cycle 3 for all trials.
Additionally, there was no significant main effect across trials for blood lactate concentration (p = 0.80) but a significant main effect among cycles (p < 0.01; η2 = 0.50; N − B = 0.99). Post hoc measures revealed that cycle 1 (8.7 ± 2.9 mmol) values were significantly lower (p < 0.01) than cycle 2 (9.6 ± 2.8 mmol) and cycle 3 (10.1 ± 3.2 mmol) and cycle 2 values also significantly lower (p = 0.05) than cycle 3 for all trials.
A 1-way repeated measure ANOVA revealed a significant main effect of trial on s-RPE (p = 0.02; η2 = 0.22; N − B = 0.74) and is shown in Figure 4. Post hoc measures showed that trial 1 was significantly lower than trial 2 (p < 0.01) and trial 3 (p = 0.05) but not trial 4 (p = 0.13). There were no significant differences found among S-RPE values among any of the other trials (i.e., T2 − T4).
Ensuring optimal recovery between exercise or training sessions should be considered an integral component for optimizing adaptations because of any training regimen (1,2,10,12,13-15). Recently, Bishop et al. (1) have acknowledged the limited knowledge regarding specific measures, modalities, and durations yielding optimal recovery in day-to-day training regimens. Further extending this problem is the lack of a dependable, noninvasive method of assessing individual recovery between sessions. Therefore, the primary aim of this study was to develop and test the efficacy of a PRS scale, an instrument designed to permit an individual to subjectively estimate their level of recovery relative to how well they feel they would perform subsequent training sessions after variable recovery periods. The major finding of the investigation was that the PRS failed to meet the a priori level of significance; however, individuals were fundamentally able to identify testing sessions yielding improved or declined repeated sprint performance (relative to the previous bout) using the PRS scale with reasonable accuracy.
To assess the utility of the PRS scale as an indicator of an individual's recovery status after a bout of repeated maximal sprint exercise, a practical statistical and interpretative approach was taken. Consequently, the primary goal of conducting the study and the specific statistical analyses was identifying PRS values that would allow individuals involved with training to approximate performance relative to their PRS before an identical training session. An approach that considered each individual's sprint performance throughout the entire testing protocol coinciding with their self-reported recovery status (i.e., PRS estimation) allowed for the greatest amount of inquiry and interpretation. Results from a chi-square analysis and the data shown in Figure 4 and Table 4 suggest that the PRS appears to be a useful, noninvasive tool that individuals are able to use to identify testing or training sessions in which they will perform better (PRS > 5) 76% of time or cases in which the individuals feel they will perform worse (PRS < 5) than their previous session about 86% of the time. That is, individual PRS ratings permitted subjects the ability to effectively determine before initiating exercise whether their recovery would permit improved performance or decreased performance. Based on these observations, it is likely that meaningful information regarding an athlete's recovery status can be gleaned from using a simple ratings scale. It is appropriate to note that this study investigated only a singular paradigm of performance (i.e., high-intensity intermittent exercise). Thus, the application of the PRS scale should be verified in other modes of exercise (e.g., endurance based).
The novel nature of the PRS scale is that it is a tool analogous to the often used and widely accepted RPE scale (24). Ratings of perceived exertion, by design, are integrative in nature as they are psychobiological manifestations that allow coaches, athletes, and researchers to monitor the amount of physiological strain and psychological discomfort experienced throughout a bout of exercise or training session (5-8). Similarly, the use of a S-RPE is also attractive in that it does not require multiple assessments of RPE during a bout, rather, one global rating of difficulty regarding the entire exercise session (3,8). However, the use of S-RPE seems impractical to monitor for signs of overtraining because of the measure being assessed after a bout of exercise. Thus, the potential utility of the PRS scale aiding in the prevention and identification of the overtraining syndrome seems promising when examined in the context of the current study.
In the current study, the PRS demonstrated exceptional value toward identifying trials in which subjects felt they were underrecovered (i.e., PRS < 5) and, subsequently, demonstrated diminished sprint performance as well as when they felt they had reached an adequate level of recovery insofar as their ability to successfully perform the testing session. Although the correlation coefficient examining the relationship between change in performance and PRS scores reached only moderate strength (−0.63) and failed to meet the a priori −0.70 level of significance, there was an obvious trend demonstrating associated increases and decreases in performance with PRS scores the majority of the time (∼80% of all trials). Moreover, individuals were best able to identify trials in which they felt underrecovered or fatigued and, subsequently, diminished performance relative to their prior bout was observed. Of major importance is that underrecovered athletes are aware they are not fully recovered and accurately reflect their inadequate recovery based on subjective feelings that can be effectively reported using the PRS scale.
This finding supports the notion that recovery is an integrative sensation that considers not only psychological state (i.e., feeling fatigued, lack of energy, etc.) but also considers the current metabolic and physiologic state before beginning the training session (2,11,18). In fact, the observation that PRS ratings effectively predicted individualized performance suggests that physiological feedback may mediate perceptually based recovery estimations. Indeed, the average performance decrement (i.e., change in sprint time) when individuals reported a PRS value less than 5 was 1.6 seconds with a maximum decline of sprint time by 6.3 seconds (∼4.5% of that individual's total sprint time).
The benefits associated with knowledge of day-to-day recovery status of athletes are obvious. As previously stated, a primary tenet to the exercise prescription process is the principle of overload. Although this principle is universally accepted, there exists a paucity of knowledge regarding the time course for optimal recovery (1). Moreover, the imbalance between planned overloading (i.e., functional overreaching) in the absence of an optimal recovery period can lead to potentially harmful overload and, ultimately, the overtraining syndrome (9,16). Although there are certainly multiple (and most likely synergistic) factors that are implicated in the development of overtraining, incomplete recovery periods after periods of overreaching is most likely the dominant influence. The diagnosis of overtraining is primarily a tedious process of excluding other maladies (16) typically confirmed only after the athlete has reached a critical level of dysfunction. Also important, recovery may show great interindividual variability. This PRS scale may prove attractive particularly for monitoring multiple athletes who may undergo similar training. With variation in recovery from 1 session to the next, the individualized nature of the PRS may permit coaches the ability to identify athletes who have or have not fully recovered.
Morgan et al. (19) reported increases in mood disruption with concomitant increases in training loads in competitive swimmers over a 10-year period and ultimately concluded psychological indices of training stressors provided better identification of signs leading to overtraining rather than monitoring physiological markers alone. Accordingly, it is generally agreed that individuals will present psychological disruption and performance decrement as early markers associated with overtraining (9,11,12,16,23). Effective assessment of inadequate recovery through use of the PRS scale would permit alteration of the pending exercise bout and in theory allow the athlete to avoid overtraining.
It is important to note that the current study included a series of 4 trials that were performed over a span of 6 days and was not necessarily designed to investigate the PRS as tool to monitor or prevent overtraining. Additionally, the mode of exercise performed in this initial study does not ensure that results reported here will transfer to other modes of exercise. However, the present investigation offers promise for effective monitoring of day-to-day recovery status related to performance potential during repeated sprint work using the PRS scale. Future work regarding the utility of the PRS scale in the prevention or diagnosis of overtraining is certainly needed for a variety of exercise situations.
Because the PRS scale was developed analogously to the S-RPE scale, it could be expected that values obtained from the PRS scale share a similar relationship with physiologic response variables typically reported with respective changes of S-RPE. That is, a perceptual response, whether S-RPE or PRS, should respond parallel to the degree of internal disruption (i.e., increased [La], HR, etc.) as a result of increased intensity or workload and may also reflect level of recovery (8). Results from this study revealed no significant difference in either HR or [La] despite changes in PRS values and significantly different S-RPE in day-to-day sprint performance. Although the lack of differences between trials for [La] and HR was not surprising because of the counterbalanced design of the recovery trials, the significant differences among S-RPEs among trials was somewhat unexpected. Recently, Green et al. (8) found no main effect for [La] or HR after a bout of constant vs. interval cycling in a thermoneutral temperature (there was a difference during a hot environment) despite a significantly different S-RPE value. Although the Green et al. study (8) compared S-RPE during 2 different modes (i.e., constant vs. interval), it is interesting that S-RPE was variable despite no change in total work completed and without significant changes in physiologic detriment (i.e., similar [La] and HR) as seen in the current investigation. It is important to note, however, that only S-RPE values after the first trial (i.e., baseline) resulted in significant differences between the recovery trials. Moreover, values obtained using the PRS scale before beginning an exercise bout mirrored the perceived level of global difficulty experienced during the trial. Indeed, the average reported value of an individual's PRS decreased (indicating lower level of perceived recovery), whereas there was a concomitant increase in S-RPE after the bout of sprints. This relationship remained consistent across all 3 recovery trials.
This study examined the potential efficacy of a newly developed PRS scale and its utility in monitoring day-to-day changes in individuals' level of perceived recovery relative to performance changes in subsequent exercise sessions of repeated intermittent sprint work. Overall, the PRS demonstrated a moderate negative correlation between level of perceived recovery and change in sprint time. Although the strength of the correlation failed to meet the a priori level of significance (r = −0.70), a detailed examination of trends within the individual data revealed that PRS ratings permitted athletes to effectively determine changes in performance relative to perceived state of recovery about 80% of the time with greater consistency shown for more extreme values (i.e., PRS score further from 5). More importantly, the scale demonstrated useful accuracy in identifying individuals showing decreased performance when reporting a feeling of being underrecovered when compared with individuals feeling at least moderately recovered and producing improved performance (86 and 77%, respectively). Perceived recovery status estimations are advantageous in that ratings are attainable before a daily training session that consequently allows appropriate adjustments in training intensity or volume to adjust for recovery status.
The ability of the PRS scale to identify changes in performance between trials of repeated sprint work in the absence of significant differences among other perceptual, physiological, and performance markers suggests that the PRS may be useful in identifying early signs of overtraining before the presentation of other established symptoms of the syndrome.
The potential utility of the PRS scale as a tool to identify early marker of overtraining should be attractive to strength and conditioning professionals and other exercise science professionals as a noninvasive, expeditious method to accurately monitor an individual's recovery status. The ability to ascertain an individuals' level of recovery before performing a subsequent bout of training is potentially useful because this would allow individual's to adjust training loads and volumes to create an optimal level of overload, thus ensuring positive training adaptations while concomitantly limiting the chance of overtraining or nonfunctional overreaching.
Certainly, more work assessing the PRS during alternative modes of exercise (i.e., aerobic training, resistance training, etc.) and more longitudinal studies examining the utility of the PRS as a tool to titrate day-to-day workloads during training or as a tool to monitor onset of overtraining are warranted.
1. Bishop, PA, Jones, E, and Woods, AK. Recovery from training: a brief review. J Strength Cond Res
22: 1015-1024, 2008.
2. Coutts, AJ, Slattery, KM, and Wallace, LK. Practical tests for monitoring performance, fatigue, and recovery in triathletes. J Sci Med Sport
10: 372-381, 2007.
3. Foster, C, Florhaug, JA, Franklin, J, Gottschall, L, Hrovativn, LA, Parker, S, Doleshal, P, and Dodge, C. A new approach to monitoring exercise training. J Strength Cond Res
15: 109-115, 2001.
4. Gomez, AL, Radzwich, RJ, Denengar, CR, Volek, JS, Rubin, MR, and Bush, JA. The effects of a 10-kilometer run on muscle strength and power. J Strength Cond Res
16: 184-191, 2002.
5. Green, JM, Crews, TR, Bosak, AM, and Peveler, WW. Overall and differentiated ratings of perceived exertion at the respiratory compensation threshold: effects of gender and mode. Eur J Appl Physiol
89: 445-450, 2003.
6. Green, JM, McLester, JR, Crews, TR, Wickwire, PJ, Pritchett, RC, and Lomax, RG. RPE association with lactate and heart rate during high-intensity interval cycling. Med Sci Sports Exerc
38: 167-172, 2006.
7. Green, JM, McLester, JR, Crews, TR, Wickwire, PJ, Pritchett, RC, and Redden, A. RPE-lactate dissociation during extended cycling. Eur J Appl Physiol
94: 145-150, 2005.
8. Green, JM, Yang, Z, Laurent, CM, Davis, JK, Kerr, K, Pritchett, RC, and Bishop, PA. Session RPE following interval and constant-resistance cycling in hot and cool environments. Med Sci Sports Exerc
39: 2051-2057, 2007.
9. Halson, SL and Jeukendrup, AE. Does overtraining exist? An analysis of overreaching and overtraining research. Sports Med
34: 967-981, 2004.
10. Jones, EJ, Bishop, PA, Richardson, MT, and Smith, JF. Stability of a practical measure of recovery from resistance training. J Strength Cond Res
20: 756-759, 2006.
11. Kentta, G and Hassmen, P. Overtraining and recovery: A conceptual model. Sports Med
1998: 1-16, 1998.
12. Lambert, M and Borresen, J. A theoretical basis of monitoring fatigue: a practical approach for coaches. Int J Sports Sci Coach
1: 371-388, 2006.
13. Lane, KN and Wenger, HA. Effect of selected recovery conditions on performance of repeated bouts of intermittent cycling separated by 24 hours. J Strength Cond Res
18: 855-860, 2004.
14. Laursen, PB and Jenkins, DG. The scientific basis for high-intensity interval training: optimising training programmes and maximising performance in highly trained endurance athletes. Sports Med
32: 53-73, 2002.
15. McLester, JR, Bishop, PA, Smith, J, Wyers, L, Dale, B, and Kozusko, J. A series of studies-A practical protocol for testing muscular endurance recovery. J Strength Cond Res
17: 259-273, 2003.
16. Meeusen, R, Duclso, M, Gleeson, M, Rietjens, G, Steinacker, J, and Urhasuen, A. Prevention, diagnosis and treatment of the overtraining syndrome: ECSS position statement 'task force'. Eur J Sports Sci
6: 1-14, 2006.
17. Mihevic, PM. Sensory cues for perceived exertion: A review. Med Sci Sports Exerc
12: 150-163, 1981.
18. Morgan, WP. Psychological components of effort sense. Med Sci Sports Exerc
26: 1071-1077, 1994.
19. Morgan, WP, Brown, DR, Raglin, JS, O'Connor, PJ, and Ellickson, KA. Psychological monitoring of overtraining and staleness. Br J Sports Med
21: 107-114, 1987.
20. Pollock, ML, Schmidt, DH, and Jackson, AS. Measurement
of cardiorespiratory fitness and body composition in the clinical setting. Clin Ther
6: 12-27, 1980.
21. Robertson, RJ and Noble, BJ. Perceptions of physical exertion; methods, mediators, and applications. Ex Sport Sci Rev
25: 407-452, 1997.
22. Sayers, SP and Clarkson, PM. Force recovery after eccentric exercise in males and females. Eur J Appl Physiol
84: 122-126, 2001.
23. Urhausen, A and Kindermann, W. Diagnosis of overtraining: What tools do we have? Sports Med
32: 95-102, 2002.
24. Utter, AC, Robertson, RJ, Green, JM, Suminski, RR, McAnulty, SR, and Nieman, DC. Validation of the adult OMNI scale of perceived exertion for walking/running exercise. Med Sci Sports Exerc
36: 416-420, 2004.