A COMPARISON OF WORKLOAD QUANTIFICATION

METHODS IN RELATION TO PHYSIOLOGICAL

RESPONSES TO RESISTANCE EXERCISE
KYLE M. GENNER AND MATTHEW WESTON
Department of Sport & Exercise Sciences, School of Social Sciences & Law, Teesside University, Middlesbrough,
United Kingdom

ABSTRACT INTRODUCTION

I
Genner, KM and Weston, M. A comparison of workload t is widely acknowledged that internal physiological
quantification methods in relation to physiological responses to responses induced primarily from training but also
resistance exercise. J Strength Cond Res 28(9): 2621–2627, from nontraining variables are responsible for the total
2014—The purpose of this study was to (a) assess the useful- physiological stress placed on athletes (23,37,38).
Referred to as internal load (2,28), this total physiological
ness of volume load (VL), session rate of perceived exertion
stress is viewed as being the determinant of whether an
(SRPE), RPE load (RPEL), and a modified RPEL (RPEL-2) to
athlete’s adaptations are positive or negative, and whether
estimate internal load from resistance exercise (RE) and (b) fur-
the outcome is performance enhancement, overtraining, ill-
ther assess the interactions between SRPE, VL, and RE inten- ness, or injury (25,18). This makes the quantification and
sity. Twelve healthy men (25 6 4 years) completed RE sessions monitoring of internal load vital for ensuring appropriate
at 55, 70, and 85 1 repetition maximum (1RM). Volume load, manipulation of training to induce desirable responses and
SRPE, RPEL, and RPEL-2 for each session were calculated, adaptations (16,23). Despite this, training is often monitored
compared, and correlated with change values (D) for blood lac- and quantified using outcome measures such as distance
tate and salivary cortisol. There were substantial increases in all traveled or the total volume of load lifted with such variables
measures of training load with progressive decreases in %1RM. often referred to as external load (45,31). Although external
There were clear substantial increases in D lactate and D cortisol load clearly provides important information (45), it may not
after RE at 55% 1RM when compared with 70 and 85%. Within- accurately reflect internal load therefore making the predic-
tion and modification of physiological outcome and adapta-
subject correlations with D cortisol were small with SRPE (r =
tion very difficult.
0.25; 90% confidence limits; 60.32), RPEL (r = 0.23; 60.32),
Although methods of measuring training load from an
RPEL-2 (r = 0.19; 60.32), and trivial for VL (r = 0.01; 60.28).
external perspective (e.g., global positioning, volume load
Correlations with D lactate were moderate with VL (r =
[VL]) and internal perspective (e.g., heart rate, blood lactate,
0.42; 60.29) and RPEL-2 (r = 0.38; 60.29), and small with hormones) are plentiful, a method to monitor internal load
SRPE (r = 0.25; 60.32) and RPEL (r = 0.25; 60.32). Correla- across all training modes and intensities is desirable. One such
tion between SRPE and VL was large (r = 0.55; 60.25). method reported across the literature (9,15,36,44) is RPE load
Although D lactate and D cortisol did not follow the same trends (RPEL), which multiplies session rate of perceived exertion
as measures of workload, VL may be superior to estimate internal (SRPE), as a measure of training intensity, with session duration,
load from RE, particularly when measured through D lactate. as a measure of training volume, to provide a training load value.
When viewing training load globally, RPEL-2 may offer the great- Previous research has validated RPEL for quantifying internal
est advantage. Finally, our results suggest that SRPE seems to load across a variety of training modes and sports (9,15,36,44).
be more closely related to VL than %1RM. Despite this, researchers are yet to fully agree on the efficacy of
SRPE to estimate resistance exercise (RE) intensity, with some
KEY WORDS training load, internal load, session RPE, RPE claiming that SRPE may actually be more closely related to the
load, global load VL of an RE session (11,31,35,41,46,48). Additionally, the use of
RPEL to quantify internal load for RE has yet to be fully as-
sessed (11,35,46,48), with the external measure of VL often pre-
Address correspondence to Kyle M. Genner, [email protected]. ferred. Although VL is an excellent method to monitor external
28(9)/2621–2627 load (33,19), there is little evidence of its effectiveness to esti-
Journal of Strength and Conditioning Research mate internal load. Additionally, the VL equation does not
Ó 2014 National Strength and Conditioning Association allow it to be used globally across all modes of training.

VOLUME 28 | NUMBER 9 | SEPTEMBER 2014 | 2621

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Workload Quantification in Resistance Exercise

As HR-based methods may not accurately reflect internal and signed informed consent forms. Ethical approval was
load during RE (16), it is suggested that SRPE, RPEL, and VL granted by Teesside University Ethics Committee.
should be assessed against the physiological responses associ-
Procedures
ated with intensity and workload in RE. Blood lactate has
One week before starting RE protocols, participants com-
previously been linked to intensity (%1 repetition maximum
pleted 5RM testing in line with recommended procedure (4),
[1RM]) and total work across various training modes, includ-
for each RE used Table 1. Five repetition maximum testing
ing RE (6,34,40). Cortisol is also altered by various RE training
was used in preference to 1RM testing to account for differ-
modes and strongly associated with changes in intensity, vol-
ences in training experience between participants, therefore
ume, and workload (1,10,24,35,42,47,51) and may also reflect
reducing safety concerns and the impact of participant anx-
stress accumulated outside of training (8,14,20,29). Therefore,
iety on test results. The tests were performed in an alternate
lactate and cortisol are appropriate variables against which to
lower-body/upper-body fashion to reduce the impact of
compare internal workload quantification methods.
acute local cumulative fatigue on test results. Test order
Because RE is programmed in terms of sets, repetitions,
was back squat, chest press, deadlift, bent over row, and
and rest intervals, and since rest intervals in RE are often
Romanian deadlift. All tests were preceded by a standardized
extremely large in comparison to actual work, the use of
warm-up, first, performing a set of 10 repetitions approxi-
total session duration to quantify training volume may not be
mately 50% of estimated 5RM. Participants then completed
appropriate (35). Consequently, the use of a modified version
10 repetitions and 7 repetitions at 70 and 90% of estimated
of RPEL, which discards the summation of rest intervals,
5RM, respectively. Testing was then completed within 3–5
therefore accounting only for the total duration of time that
attempts per exercise, using increases of 5–10% if the pre-
work was actually being performed (modified RPEL
vious set was completed successfully. A set was successfully
[RPEL-2]), warrants assessment.
completed when all repetitions were completed with the
The aims of this study were therefore to (a) assess the
appropriate technique. Where a set was not successfully
usefulness of VL, SRPE, RPEL, and RPEL-2 to estimate
completed, the load was reduced by 2.5–5% for the subse-
internal load from RE and (b) further assess the interactions
quent set. A rest period of 3 minutes was given between each
between SRPE, VL, and RE intensity.
attempt. Technique standardization for each exercise was in
line with recommendations (4), with the exception of the
METHODS
Romanian deadlift. Technique for this exercise was standard-
Experimental Approach to the Problem ized using the following criteria: start with feet between hip
Subjects each completed 3 RE protocols using 85, 70, and and shoulder width apart; keep chest up, back straight, and
55% 1RM in a counterbalanced, randomized crossover slight bend in knees; keep bar close to body lowering barbell
design. Volume load, SRPE, RPEL, and RPEL-2 were to just below kneecap; and keeping back straight, stand up,
subsequently calculated to allow cross method comparison and fully extend hips. This technique was used because of
across all RE protocols. For each protocol, subjects provided safety concerns regarding the lumbar spine when lowering
blood lactate and salivary cortisol samples immediately the barbell to the floor without appropriate posterior chain
before, immediately after, and 30 minutes post-RE. These flexibility. Individual loads for each session and exercise were
values were subsequently used to measure the relationships calculated based on 1RM’s estimated from the 5RM tests (4).
between the workload quantification methods and the Participants were also familiarized with blood lactate and
internal physiological response.

Subjects
Twelve physically active men (aged 25 6 4 years; age range
19–30 years; height: 180 6 7 cm; mass: 77 6 10 kg), with at
TABLE 1. Strength testing data (mean 6 SD) all
least 1-year RE experience, were recruited to participate in
exercises used.*
this study. All reported participating in an RE program of
at least 2 sessions per week for the 12 months before the Estimated 1RM
study. All were also asked to conform to the following in- 5RM (kg) (kg)
structions for each protocol: consume the same diet in the
Back squat 97.29 6 19.47 112.08 6 22.81
24 hours before participation, refrain from eating 3–4 hours Chest press 77.29 6 12.94 89.17 6 14.75
before participation, consume only water during and for Deadlift 110.42 6 25.40 126.88 6 29.33
30 minutes after participation, and abstain from exercise, Bent over row 68.54 6 11.00 78.86 6 13.29
alcohol, and caffeine for at least 48 hours before each session. Romanian 88.75 6 11.51 101.82 6 13.83
Participants confirmed that they had adhered to the instruc- deadlift
tions before commencing each session, although diet records *RM = repetition maximum.
were not taken. Subjects were screened using a medical
questionnaire to confirm adequate health, and all completed
the TM

2622 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

TABLE 2. Descriptive data (mean 6 SD) for the 4 measures of training load and the D cortisol and D lactate following
the 3 resistance training protocols (55, 70, and 85% 1RM).*

Protocol VL (kg) SRPE RPEL RPEL-2 D cortisol (mg$dl21) D lactate (mmol$L21)

55% 1RM 12,396 6 944 8.0 6 1.6 4596 6 1145 347 6 50 0.33 6 0.39 5.7 6 2.0
70% 1RM 10,560 6 1753 6.9 6 1.4 3154 6 589 279 6 70 0.05 6 0.19 4.8 6 1.7
85% 1RM 8319 6 1412 6.2 6 2.2 2168 6 715 248 6 77 0.25 6 0.19 5.1 6 2.1

*1RM = 1 repetition maximum; VL = volume load; SRPE = session rate of perceived exertion; RPEL = RPE load; RPEL-2 =
modified RPEL.

salivary cortisol sampling techniques, the SRPE scale. Statistical Analyses

Because of participant, facility, and equipment availability, par- Data are presented as the mean 6 SD. Using a custom-made
ticipants worked in 2 subgroups training from 19.00 to 21.00, spreadsheet (21), all data were log transformed and then
and 09.00 to 11.00, respectively. Each participant trained at the back transformed to obtain the percent difference, with
same time of day for each session to account for the circadian uncertainty of the estimates expressed as 90% confidence
rhythm of cortisol (49). After a standardized warm-up, each limits (CLs), between training loads (VL, SRPE, RPEL,
session was completed using 3 sets of each exercise performed and RPEL-2) for the 3 resistance training protocols (55,
in the same order, with 2 minutes rest between sets. As Fisher 70, and 85% 1RM). This is the appropriate method for quan-
et al. (13) asserted that intensity in its truest sense refers to the tifying changes in athletic performance (22). Inference was
level of effort applied at a given load, it is argued that studies then based on the disposition of the confidence interval for
assessing workload at different percentages of 1RM should the mean difference to the smallest worthwhile effect; the
control set end point according to levels of relative effort at probability (percent chances) that the true population dif-
each given workload. To that end, a set was complete when ference between RE sessions was substantial (.0.2 SDs) or
the participant perceived that they were 1–2 repetitions from trivial was calculated as per the magnitude-based inference
volitional exhaustion. Protocols were performed with 7 days of approach (5). These percent chances were qualified
recovery to minimize the impact of cumulative fatigue. through probabilistic terms assigned using the following
Blood was collected through finger venepuncture and scale: ,0.5%, most unlikely; 0.5–5%, very unlikely; 5–25%,
analyzed for lactate using a YSI 2300 analyzer (YSI UK ltd, unlikely; 25–75%, possibly; 75–95%, likely; 95–99.5%, very
Hampshire, United Kingdom). Saliva was taken through likely; .99.5%, most likely (22). A within-subject design
Sarstedt salivette and analyzed for cortisol using Salimetrics was used to determine if high workload load scores (VL,
high-sensitivity salivary cortisol enzyme immunoassay kit SRPE, RPEL, and RPEL-2) were associated with greater
and a Labsystems Multiscan Microplate Reader (Labsys- changes in postsession cortisol and blood lactate. This is
tems, Helsinki, Finlad). Values were subsequently used to the appropriate method as it permits the analysis of within-
calculate a delta (change; D) score for each participant for subject changes by removing between-subject differences
lactate and cortisol by subtracting the prevalues from the (7). Within-subject correlations were therefore calculated
greatest of the post and post-30 values. Salivary cortisol between the participants’ VL and SRPE scores, and D cor-
has been found to be closely related to serum unbound cor- tisol and D lactate with the measures training load follow-
tisol (r = 0.93) (39), with previous studies reporting its high- ing the 3 resistance training protocols (55, 70, and 85%
est point of elevation to be at 30 minutes before exercise 1RM). Confidence limits (90%) for the within-participant
correlations were calculated as per Altman and Bland (3).
(50,47). To ensure reliability of measures, intratester test-
The following scale of magnitudes (22) was used to inter-
retest reliability was completed using samples obtained from
pret the magnitude of the correlation coefficients: ,0.1,
the 12 participants during 5RM testing. Results were ana-
trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large;
lyzed using coefficient of variation (CV) and intraclass cor-
0.7–0.9, very large; .0.9, nearly perfect.
relation (ICC) with CV calculated as 5.6% for salivary
cortisol and 2.1% for blood lactate, and ICC values of 0.98
and 0.90, respectively. RESULTS
Session RPE was recorded at post-30 for each session by Descriptive data for the 3 resistance training protocols are
asking participants “how was your workout” (16), whereas presented in Table 1. We observed clear substantial increases
time per set and reps completed were recorded throughout in all measures of training load (VL, SRPE, RPEL, and
each session by a randomly assigned training partner. Values RPEL-2) with progressive decreases in exercise intensity,
were then used to calculate RPEL, RPEL-2, and VL (19). as determined by % of 1RM (Table 2). Clear substantial

VOLUME 28 | NUMBER 9 | SEPTEMBER 2014 | 2623

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Workload Quantification in Resistance Exercise

TABLE 3. Percentage difference with 90% CL and practical inferences, for between protocol comparisons (55, 70,
and 85% 1RM) on outcome measures.*

Intensities (%1RM) Mean % difference (690% CL) Practical inference

VL 55–70% 17 6 63 Most likely a substantial difference

55–85% 46 6 79 Most likely a substantial difference
70–85% 24 6 7.5 Most likely a substantial difference
SRPE 55–70% 16 6 9.4 Very likely a substantial difference
55–85% 29 6 12 Most likely a substantial difference
70–85% 12 6 17 Likely a substantial difference
RPEL 55–70% 21 6 11 Very likely a substantial difference
55–85% 36 6 12 Most likely a substantial difference
70–85% 12 6 17 Likely a substantial difference
RPEL-2 55–70% 49 6 15 Most likely a substantial difference
55–85% 107 6 14 Most likely a substantial difference
70–85% 39 6 21 Very likely a substantial difference
D lactate (mmol$L21) 55–70% 18 6 16 Likely a substantial difference
55–85% 15 6 13 Likely a substantial difference
70–85% 23 6 17 Possibly a trivial difference
D cortisol (mg$dl21) 55–70% 148 6 156 Very likely a substantial difference
55–85% 102 6 184 Likely a substantial difference
70–85% 219 6 44 Possibly a substantial difference

*CL = confidence limit; 1RM = 1 repetition maximum; VL = volume load; SRPE = session rate of perceived exertion; RPEL = RPE
load; RPEL-2 = modified RPEL.

increases were also recorded for the D lactate and D cortisol compared with 70 and 85% 1RM, although the lack of
following the 55% 1RM protocol when compared with the 70 substantial difference between RE at 70 and 85% shows that
and 85% 1RM protocols (Table 3). The differences for the D internal responses did not follow the same trends as
lactate and D cortisol following the 70% 1RM protocol when measures of workload. The findings also suggest that
compared with the 85% 1RM protocols were less clear. measures of workload, particularly VL and RPEL-2, may
There were small correlations between SRPE (r = 0.25; be related to internal load when measured through change
90% CLs; 60.32), RPEL (r = 0.23; 60.32), and RPEL-2 (r = in lactate, but not when measured by change in salivary
0.19; 60.32) with D cortisol and a trivial correlation between cortisol. Finally, our results showed that SRPE did not reflect
VL with D cortisol (r = 0.01; 60.28). Moderate correlations intensity when expressed as %1RM, although it was largely
were shown between VL (r = 0.42; 60.29) and RPEL-2 (r = associated with VL.
0.38; 60.29) with D lactate, and small correlations between Greater workloads were achieved at lower intensities and
SRPE (r = 0.25; 60.32) and RPEL (r = 0.25; 60.32) with D lower workloads at higher intensities, for all measures of
lactate. There was also a large correlation between SRPE workload. This suggests that with set end point and interset
and VL (r = 0.55; 60.25). rest standardized, RE at a lower %1RM may result in greater
internal loads. This was somewhat supported by the changes
DISCUSSION in cortisol and lactate, which were also greatest at 55% 1RM.
The aims of this study were to (a) assess the usefulness of In contrast to measures of workload, the less clear differ-
VL, SRPE, RPEL, and RPEL-2 to estimate internal load ences in cortisol and lactate change from RE at 70% 1RM
from RE and (b) further assess the interactions between compared with 85% 1RM make it difficult to make any
SRPE, VL, and RE intensity. This is the first study to assess strong inferences about the differences in physiological
the use of SRPE, VL, RPEL, and RPEL-2 to estimate response between these sessions. This perhaps suggests that
internal load from RE at different intensities and further adds the between-intensity differences in measures of workload
to the literature regarding the interaction between SRPE, may have been contributed to by physiological variables not
VL, and RE intensity. All measures of workload substantially measured in the present study. This observation is supported
differentiated between RE at different intensities, with by the small and trivial correlations between measures of
greater workload values at 55% 1RM and lower workload workload and cortisol. The moderate positive correlations
values at 85% 1RM. Our participants also exhibited sub- seen from VL and RPEL-2 with lactate suggested that these
stantially higher internal loads from RE at 55% 1RM measures of workload may moderately reflect internal load
the TM

2624 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

from RE at different intensities. The small correlations seen RE, previous research has reported RPE to be related to
from SRPE and RPEL with lactate suggest these may be less neuromotor activity as measured by electromyography
able to reflect internal load from RE at different intensities. (26,27). Additionally, other researchers have reported a vari-
Because of the lack of previous research assessing the ety of other neuromuscular, metabolic, and hormonal meas-
usefulness of VL, RPEL, and RPEL-2 to quantify workload ures to be altered from RE (10,27,34,42,47). It is therefore
from RE, comparisons to the current study are difficult. possible that measuring alternative physiological variables
McBride et al. assessed the use of various workload may have resulted in greater relationships with our measures
quantification methods from hypertrophy, strength, and of workload, allowing a greater insight into their effective-
power protocols, although inferences about their ability to ness as measures estimators of internal load.
estimate internal load could not be made as none of these The relationship between VL and lactate seemed similar
methods were compared with any measure of internal to those reported by previous researchers (34,47), with lac-
physiological response. Regardless, the authors reported tate response suggested to largely be a product of the total
greater VL values as a result of the strength protocol, work encountered during RE. Although previous research
although this could be explained by the large differences in has failed to assess the relationship of the lactate response to
sets and reps performed between protocols. Additionally, RE with SRPE, RPEL, and RPEL-2, the results of the cur-
interset rest period and set end point relative to effort were rent study suggested that RPEL-2 may be the superior of
not accounted for, meaning the results cannot truly be used these 3 measures, closely behind VL. This is possibly
to assess the ability of VL to differentiate between RE at because of its ability to account for the total time working
different intensities. Our results were in-line those of rather than the full session duration, which may be an
Pritchett et al. who reported that VL was greater from important consideration when assessing training protocols
a session performed at 60% compared with 90% 1RM. The with long rest periods (35). As the correlation was close to
authors standardized set end point, having all participants that observed between lactate and VL, RPEL-2 may be sim-
perform sets to volitional exhaustion. They also standardized ilarly capable of estimating internal load, with the advantage
interset rest period, therefore reducing the impact of further of also being usable in other training modes. This suggests
confounding variables on the results. that RPEL-2 may provide a useful global workload quanti-
Previous research has reported that in RE lactate and fication method, although the moderate correlation again
cortisol responses tend to be greater from protocols using suggests that other physiological variables may have also
the greatest VL, lowest %1RM, and the shortest rest periods contributed to the RPEL-2 values.
(10,34,47). We report similar findings, although perhaps Session RPE did not reflect intensity when expressed as %
because of the standardized interset rest periods and stan- 1RM, although it did have a large relationship with VL. The
dardized set end point, smaller between intensity differences results of the current study are supported by Pritchett et al.
were evident in the current study. Additionally, the differ- who reported that SRPE was largely associated with VL,
ence in lactate and cortisol responses between RE at 70 and even when the higher VL values were achieved from session
85% seemed smaller and less clear in the current study than performed at the lowest %1RM. Lodo et al. further
previously reported. The current study’s standardization of supported these findings with a 2 part experimental design.
set end point and interset rest period during the current In the first experiment, participants each performed RE in
study again may be able to account for these inconsistencies. a strength-orientated session (10 sets of 4RM), a hypertrophy
The lack of strength in correlations between measures of session (8 sets of 8RM), and an endurance session. Signif-
workload and salivary cortisol suggests that these measures icant positive correlations were found between VL and
do not accurately reflect cortisol response to RE. Similar SRPE with the highest values achieved from RE at higher
findings were reported by McGuigan et al., although only intensities. In the second experiment, participants each
between SRPE and cortisol. Previous research has also performed RE sessions at 50 and 75% 1RM with VL
shown that cortisol responses from RE are greater in fitter matched between sessions. Significant positive correlations
individuals (32,17). It is therefore possible that the heteroge- were again found between SRPE and VL, and no significant
neity in strength levels observed in the present study may differences were found between intensities for SRPE.
have been an important influencing factor on the relation- Despite this, other studies have reported that SRPE is
ship between salivary cortisol and measures of workload. a valid and reliable method to monitor RE intensity and
(Table 1). As SRPE and RPEL have previously been used different types of RE, with SRPE tending to reflect intensity
to accurately reflect internal load through other physiologi- when expressed as %1RM (11,30,35,46,48). There are several
cal measures from other training modes (9,15,35,42), it is possible explanations for these differences. First, all of these
possible that salivary cortisol is not an accurate measure of studies failed to control or account for VL. Without this
internal load from RE. Previous research has reported that being controlled or accounted for it is not possible to clearly
RPE can be influenced by a host of psycho-physiological infer whether the SRPE values may have been more strongly
variables (12), possibly explaining why the relationships associated with VL than %1RM, although based on the find-
found in the current study were not stronger. Specific to ings of the current study, and the similar findings from

VOLUME 28 | NUMBER 9 | SEPTEMBER 2014 | 2625

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Workload Quantification in Resistance Exercise

Pritchett et al. and Lodo et al., this could be a possibility. the current study. As set end point in the current study was
Interestingly, retrospective analysis of the methods from regulated by individual perception, potential discrepancies in
those studies showing a relationship between %1RM and effort across participants may have affected the results.
SRPE does allow an alternative version of VL (AVL) to be Furthermore, participants were required to adjust load for
calculated for the RE sessions at all intensities performed. their partner between sets, which would likely have influ-
Alternative version of VL multiplies sets 3 reps 3 the % enced physiological responses and RPE values. As previous
1RM values for each session, rather than using the absolute research has reported the benefit of using RPEL and SRPE
load values usually used (18). Reviewing the results using across a variety of training modes, the findings of the current
these values reveals variable findings. In 2 of the 4 studies, study should not be interpreted as reason to no longer use
AVL followed the same trend as SRPE (35,30), whereas in them. Despite this, the current findings do suggest that more
the other 3 (11,46,48), AVL values were inverse to SRPE. research about their usefulness across training modes using
Despite this, further analysis of the methods reveal that in a variety of methodological approaches is warranted, with
these 3 studies, RE was performed at 90% 1RM using 4–5 a particular focus on RE.
reps, 70% 1RM using 10 reps, and 50% 1RM using 15 reps. As
previous authors have asserted that intensity in its truest sense REFERENCES
refers to the level of effort applied at a given load (13), it is of 1. Ahtiainen, J, Pakarinen, A, Kraemer, W, and Häkkinen, K. Acute
interest to review these results according to this premise. hormonal and neuromuscular responses and recovery to forced vs.
Using RM prediction tables (4) and equations (43) reveals that maximum repetitions multiple resistance exercises. Int J Sports Med
24: 410–418, 2003.
performing 4–5 reps at 90% 1RM, 10 reps at 70% 1RM, and 15
2. Alexiou, H and Coutts, A. A comparison of methods used for
reps at 50% 1RM equates to approximately 100, 85, and ,70% quantifying internal training load in women soccer players. Int
of an estimated maximal effort at each given load, respectively. J Sports Physiol Perform 3: 320–330, 2008.
Indeed, looking across all studies discussed, it can be seen that 3. Altman, D and Bland, M. How to obtain the confidence interval
when estimated relative effort was similar between sessions at from a P value. BMJ 343: 2090, 2011.
a different %1RM, VL or AVL was associated with or followed 4. Baechle, T and Earle, R. Essentials of Strength Training and
Conditioning. Champaign, IL: Human Kinetics, 2008.
a similar trend to SRPE. When VL or AVL was equal or
inverse to SRPE, estimated relative effort followed a similar 5. Batterham, A and Hopkins, W. Making meaningful inferences about
magnitudes. Int J Sports Physiol Perform 1: 50–57, 2006.
trend to SRPE. In contrast, when VL or AVL was equal or
6. Borg, G, Hassmen, P, and Lagerstrom, M. Perceived exertion related
inverse to SRPE, with set end point standardized in terms of to heart rate and blood lactate during arm and leg exercise. Eur J
relative effort, %1RM did not follow the same trends as SRPE. Appl Physiol Occup Physiol 56: 679–685, 1987.
Given these facts and that SRPE asks participants the broad 7. Bland, J and Altman, D. Calculating correlation coefficients with repeated
question “how hard was your session” rather than “how observations: Part 1-correlation within subjects. BMJ 310; 633, 1995.
intense was your session in relation to your 1RM,” it is possi- 8. Burke, H, Davis, M, Otte, C, and Mohr, D. Depression and cortisol
responses to psychological stress: A meta-analysis.
ble that VL and the level of effort applied at a given %1RM Psychoneuroendocrinology 30: 846–856, 2005.
may have a greater influence on SRPE than %1RM alone. 9. Coutts, A, Rampinini, E, Marcora, S, Castagna, C, and Impellizzeri, F.
Based on the current findings, and those of previous studies, Heart rate and blood lactate correlates of perceived exertion during
it is likely that SPRE in RE is affected by more than simply small-sided soccer games. J Sci Med Sport 12: 79–84, 2009.
%1RM with VL likely a contributory factor and relative effort 10. Crewther, B, Cronin, J, Keogh, J, and Cook, C. The salivary
testosterone and cortisol response to three loading schemes.
at a given load also potentially playing a role. Future studies
J Strength Cond Res 22: 250–255, 2008.
may look to further assess exactly how these variables interact
11. Day, M, McGuigan, M, Brice, G, and Foster, C. Monitoring
with each during RE. intensities during resistance training using a session RPE scale.
J Strength Cond Res 18: 353–358, 2004.
PRACTICAL APPLICATIONS 12. Eston, R. Use of ratings of perceived exertion in sports. Int J Sports
Although all measures of workload were able to differentiate Physiol Perform 7: 175–182, 2012.
between RT at different intensities, VL and RPEL-2 may be 13. Fisher, J, Steele, J, Bruce-Low, S, and Smith, D. Evidence based
resistance training recommendations. Med Sport 15: 147–162, 2011.
superior for estimating the internal load achieved. Therefore,
14. Follenius, M, Brandenberger, G, Oyono, S, and Candas, V. Cortisol as
when RE is the only training mode, VL may offer the best of a sensitive index of heat-intolerance. Physiol Behav 29: 509–513, 1982.
the 4 methods to monitor workloads. When other training 15. Foster, C. Monitoring training in athletes with reference to
modes are being used, RPEL-2 may be the method of overtraining syndrome. Med Sci Sports Exerc 30: 1164–1168, 1998.
choice, although more research into its use during RE and 16. Foster, C, Florhaug, J, Franklin, J, Gottschall, L, Hrovatin, L,
other training modes are required before it can be broadly Parker, S, Doleshal, P, and Hodge, C. A new approach to
recommended. It may also be of interest to practitioners and monitoring exercise training. J Strength Cond Res 15: 109–115, 2001.
researchers to monitor intensity using relative effort at 17. Fry, A, Kraemer, W, Stone, M, Koziris, L, Thrush, J, and Fleck, S.
Relationships between serum testosterone, cortisol and weightlifting
a given load rather than a percentage of maximal effort, performance. J Strength Cond Res 14: 338–343, 2000.
although this again needs more research before being 18. Fry, R, Morton, A, and Keast, D. Overtraining in athletes: An
broadly recommended. There were several limitations to update. Sports Med 12: 32–65, 1991.
the TM

2626 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

19. Haff, G. Quantifying workloads in resistance training: A brief review. 36. Milanez, V, Spiguel Lima, M, Gobatto, C, Perandini, L, Nakamura, F,
Prof Strength Conditioning 19: 4–13, 2010. and Ribeiro, L. Correlates of session-rate of perceived exertion (RPE)
20. Hansen, A, Garde, A, and Persson, R. Sources of biological and in a karate training session. Sci Sports 26: 38–43, 2011.
methodological variation in salivary cortisol and their impact on 37. Morgan, P, Brown, D, Raglin, J, O’Connor, P, and Ellickson, K.
measurement among healthy adults: A review. Scand J Clin Lab Psychological monitoring of overtraining and staleness. Br J Sports
Invest 68: 448–458, 2008. Med 21: 107–114, 1987.
21. Hopkins, W. Spreadsheets for analysis of controlled trials with 38. Morgan, P, Costill, D, Flynn, G, Raglin, J, and O’Connor, P. Mood
adjustment for a predictor. Sportscience 10: 46–50, 2006. (sportsci. disturbance following increased training in swimmers. Med Sci Sports
org/2006/wghcontrial.htm). Exerc 20: 408–414, 1998.
22. Hopkins, W, Marshall, S, Batterham, A, and Hanin, J. Progressive 39. O’Connor, P and Corrigan, D. Influence of short-term cycling on
statistics for studies in sports medicine and exercise science. Med Sci salivary cortisol levels. Med Sci Sports Exerc 19: 224–228, 1987.
Sports Exerc 41: 13–23, 2009.
40. Oliver, S, Costa, R, Laing, S, Bilzon, J, and Walsh, N. One night of
23. Impelizzeri, F, Rampinini, E, and Marcora, S. Physiological sleep deprivation decreases treadmill endurance performance. Eur J
assessment of aerobic training in soccer. J Sports Sci 23: 583–592, 2005. Appl Physiol 107: 155–161, 2009.
24. Jacks, D, Sowash, J, Anning, J, McGloughlin, T, and Andres, F. Effect 41. Pritchett, R, Green, J, Wickwire, P, Pritchett, K, and Kovacs, M. Acute
of exercise at three exercise intensities on salivary cortisol. J Strength and session RPE responses during resistance training: Bouts to failure
Cond Res 16: 286–289, 2002. at 60% and 90% of 1RM. S Afr J Sports Med 21: 23–26, 2009.
25. Kentta, G and Hassmen, P. Overtraining and recovery. Sports Med 42. Ratamess, N, Kraemer, W, Volek, J, Maresh, C, Van Heest, J,
26: 1–16, 1998. Sharman, M, Rubin, M, French, D, Vescovi, J, Silvestre, R,
26. Lagally, K, McCaw, S, Young, G, Medema, H, and Thomas, D. Hattfield, D, Fleck, S, and Deschenes, M. Androgen receptor
Ratings of perceived exertion and muscle activity during the bench content following heavy resistance exercise in men. J Steroid Biochem
press exercise in recreational and novice lifters. J Strength Cond Res Mol Biol 93: 35–42, 2005.
18: 359–364, 2004. 43. Reynolds, J, Gordon, T, and Robergs, R. Prediction of 1 repetition
27. Lagally, K, Robertson, R, Gallagher, K, Goss, F, Jakicic, J, Lephart, S, maximum strength from multiple repetition maximum testing and
McCaw, S, and Goodpaster, B. Perceived exertion, anthropometry. J Strength Cond Res 20: 584–592, 2006.
electromyography, and blood lactate during acute bouts of 44. Rodriguez-Alonso, M, Fernandez-Garcia, B, Perez-Landaluce, J, and
resistance exercise. Med Sci Sports Exerc 34: 552–559, 2002. Terrados, N. Blood lactate and heart rate during national and
28. Lambert, M and Borresen, J. Measuring training load in sports. Int J international women’s basketball. J Sports Med Phys Fitness 43: 432–
Sports Physiol Perform 5: 406–411, 2010. 436, 2003.
29. Leproult, R, Copinschi, G, Buxton, O, and Van Cauter, E. Sleep loss 45. Scott, T, Black, C, Quinn, J, and Coutts, A. Validity and reliability of
results in an elevation of cortisol levels the next evening. Sleep 20: the session-RPE method for quantifying training in Australian
865–870, 1997. football: A comparison of the CR10 and CR100 scales. J Strength
Cond Res 27: 270–276, 2013.
30. Linz-Filho, O, Robertson, R, Farah, B, Rodrigues, S, Cyrino, E, and
Ritti-Dias, R. Effects of exercise intensity on rating of perceived 46. Singh, F, Foster, C, Tod, D, and McGuigan, M. Monitoring different
exertion during a multiple-set resistance exercise session. J Strength types of resistance training using session rating of perceived
Cond Res 26: 466–472, 2012. exertion. Int J Sports Physiol Perform 2: 34–45, 2007.
31. Lodo, L, Moreira, A, Zavanela, P, Newton, M, McGuigan, R, and 47. Smillios, I, Pilianidis, T, Karamouzis, M, and Tokmakidis, S.
Aoki, M. Is there a relationship between the total volume of load Hormonal response after various resistance exercise protocols. Med
lifted in bench press exercise and the rating of perceived exertion? Sci Sports Exerc 35: 644–654, 2003.
J Sports Med Phys Fitness 52: 483–488, 2012. 48. Sweet, T, Foster, C, McGuigan, M, and Brice, G. Quantitation of
32. Marthur, D, Toriola, A, and Dada, O. Serum cortisol and resistance training using the session rating of perceived exertion
testosterone levels in conditioned male distance runners and non- method. J Strength Cond Res 18: 796–802, 2002.
athletes after maximal exercise. J Sports Med Phys Fitness 26: 242– 49. Thuma, J, Gilders, R, Verdum, M, and Louks, A. Circadian
250, 1986. rhythm of cortisol Confounds cortisol responses to exercise:
33. McBride, J, McCaulley, G, Cormie, P, Nuzzo, J, Cavill, Michael, J, Implications for future research. J Appl Physiol (1985) 78: 1657–
and Triplett, T. Comparison of methods to quantify volume during 1664, 1995.
resistance exercise. J Strength Cond Res 23: 106–110, 2009. 50. Van Bruggen, M, Hackney, A, McMurray, R, and Ondrak, K. The
34. McCaulley, G, McBride, J, Cormie, P, Hudson, M, Nuzzo, J, relationship between serum and salivary cortisol levels in response
Quindry, J, and Triplett, T. Acute hormonal and neuromuscular to different intensities of exercise. Int J Sports Physiol Perform 6: 396–
responses to hypertrophy, strength and power type resistance 407, 2011.
exercise. Eur J Appl Physiol 105: 695–704, 2009. 51. Williams, A, Ismail, N, Sharma, A, and Jones, D. Effects of
35. McGuigan, M, Egan, A, and Foster, C. Salivary cortisol responses resistance exercise volume and nutritional supplementation on
and perceived exertion during High Intensity and Low Intensity anabolic and catabolic hormones. Eur J Appl Physiol 86: 315–321,
Bouts of Resistance Exercise. J Sports Sci Med 3: 8–15, 2004. 2002.

VOLUME 28 | NUMBER 9 | SEPTEMBER 2014 | 2627

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.