Alcazar J et al. The Force-Velocity Relationship in … Int J Sports Med
Training & Testing
Thieme
Introduction
Muscle power has increasingly been shown to be an important de-
terminant of athletic performance [36]. However, the importance
of muscle power oversteps the context of sport, and early studies
observed a strong association between muscle power and several
indices of functional performance in older adults [1]. Muscle power
has a greater inuence on functional mobility in older adults than
any other physical capacity [6], and it has been recognized as a pri-
ority target of resistance training interventions aimed at enhanc-
ing physical function and preserving independence in later life [21].
The individual capacity to produce muscle power depends on the
ability to exert force and velocity, and consequently, on the force-
velocity (F-V) relationship. The evaluation of the F-V relationship in
older adults might help to identify specic neuromuscular decits
and optimize the design of exercise programs to counteract such
decits and improve physical performance, as has been reported
in young adults regarding jumping performance [25].
The evaluation of the F-V relationship usually consists of register-
ing the movement velocity exerted against increasing loads (isotonic
evaluation) or measuring the force exerted at dierent constant
The Force-Velocity Relationship in Older People: Reliability and
Validity of a Systematic Procedure
Authors
Julian Alcazar
1, 2
, Carlos Rodriguez-Lopez
1, 2
, Ignacio Ara
1, 2
, Ana Alfaro-Acha
2, 3
, Asier Mañas-Bote
1, 2
,
Amelia Guadalupe-Grau
2, 4
, Francisco Jose García-García
2, 4
, Luis M. Alegre
1, 2
Aliations
1 GENUD Toledo Research Group, Universidad de Castilla-
La Mancha, Toledo, Spain
2 CIBER of Frailty and Healthy Aging (CIBERFES), Madrid,
Spain
3 Hospital Virgen del Valle, Complejo Hospitalario de
Toledo, Toledo, Spain
4 ImFINE Research Group, Universidad Politécnica de
Madrid, Madrid, Spain
Key word
aging, muscle power, power training, resistance training,
functional ability, muscle function
accepted 26.08.2017
Bibliography
DOI https://doi.org/10.1055/s-0043-119880
Published online: 2017
Int J Sports Med
© Georg Thieme Verlag KG Stuttgart · New York
ISSN 0172-4622
Correspondence
Dr. Luis M. Alegre, PhD
University of Castilla-La Mancha
Faculty of Spors Sciences, Campus Tecnológico
Avda. Carlos III s/n
45071 Toledo
Spain
Tel.: + 34/925/268 800, e:5520, Fax: + 34/925/268 846
Luis.Alegre@uclm.es
ABSTRACT
This study compared the reliability and validity of dierent
protocols evaluating the force-velocity (F-V) relationship and
muscle power in older adults. Thirty-one older men and wom
-
en (75.8 ± 4.7 years) underwent two F-V tests by collecting the
mean and peak force and velocity data exerted against increas
-
ing loads until one repetition maximum (1RM) was achieved in
the leg press exercise. Two attempts per load were performed,
with a third attempt when F-V points deviated from the linear
F-V regression equation. Then, the subjects performed 2 × 3
repetitions at 60 % 1RM to compare purely concentric and ec
-
centric-concentric repetitions. The Short Physical Performance
Battery was conducted to assess the validity of the dierent
protocols. Signicant dierences were found in maximal pow
-
er (Pmax) between mean and peak values and between proto-
cols diering in the number of attempts per load (p < 0.01).
Registering mean values, a third attempt, and multiple loads
(>3), was significantly more reliable (Pmax: CV = 2.6 %;
ICC = 0.99) than the other alternatives. Mean values were also
observed to be more associated with physical function than
peak values (R
2
= 0.34 and 0.15, respectively; p < 0.05). No sig-
nicant dierences were observed between concentric and
eccentric-concentric repetitions. Thus, collecting mean force
and velocity values against multiple loads, while monitoring
the linearity of the F-V relationship, seemed to be the more
adequate procedure to assess the F-V prole and muscle pow
-
er in older adults.
Downloaded by: National University of Singapore. Copyrighted material.
Alcazar J et al. The Force-Velocity Relationship in … Int J Sports Med
Training & Testing
Thieme
velocities (isokinetic evaluation). The latter might present some
disadvantages for the evaluation of physical function in older
adults, because isokinetic movement rarely occurs during func-
tional tasks. Thus, the evaluation of the F-V relationship can be a
time-consuming and fatiguing task for an older subject, who has
to perform a relatively high number of repetitions against dierent
loads/velocities to draw the whole F-V relationship. Fortunately,
the F-V relationship during multi-joint movements has been shown
to follow a strong linear regression pattern [4, 23], which permits
the F-V relationship to be accurately drawn from a few F-V points
by means of a linear regression equation [39], something that could
facilitate the evaluation of the F-V relationship in older adults. How-
ever, whereas in young adults a recent study demonstrated that a
two-load method was a feasible approach to assess the F-V rela-
tionship in an upper-body resistance exercise [31], the validity and
reliability of the determination of the F-V relationship through the
evaluation of a few loads have not been previously analyzed in older
adults. In addition, some concerns remain regarding the protocol
conducted for F-V and muscle power measurement in older adults
(e. g., collecting mean vs. peak values, concentric vs. eccentric-con-
centric muscle actions, number of attempts performed with each
increasing load, and the total number of loads evaluated to build
the F-V relationship).
Thus, the main goals of this investigation were to compare the
reliability of dierent protocols that evaluate the F-V relationship,
and to assess the validity and reliability of a systematic procedure
to assess the F-V relationship and muscle power in older adults.
Methods
Participants
Subjects were recruited through advertisements and community
newsletters. Participants were screened if they were aged ≥ 70
years, community-dwelling, and reported no participation in a reg-
ular resistance training program in the previous 6 months. The sub-
jects completed a medical history questionnaire and performed
the Short Physical Performance Battery (SPPB) [17] to assess their
physical function. Exclusion criteria included a SPPB score < 4, se-
vere cognitive impairment (mini-mental state examination (MMSE)
score < 20), neuromuscular or joint injury, stroke, myocardial in-
farction or bone fracture in the previous six months, uncontrolled
hypertension ( > 200/110 mmHg) or terminal illness. A total of 31
older subjects (17 women) were given medical acceptance by the
study physician and met the entry criteria to participate in the study
(▶Table 1). All the subjects gave their informed consent and the
study was performed in accordance with the Helsinki Declaration
and approved by the Ethical Committee of the Toledo Hospital. This
study meets the ethical standards in sports and exercises science
research [18].
Testing Procedures
First, the participants attended 2 familiarization sessions. Then, leg
muscle power testing was conducted twice, on separate days, by
the same evaluator, with identical equipment and procedures, and
at the same time of the day with a dierence of 7 days in between.
F-V and muscle power testing
Before each session, subjects performed a general warm-up con-
sisting of 5 min of cycling (Ergoline, 800S, Bitz, Germany) at a self-
reported light intensity (10–40 W), plus a specic warm-up in which
the subjects performed 3 sets of 10 repetitions on the leg press
equipment (BH Fitness, L050, Vitoria, Spain) at an intensity equiva-
lent to 40 % of their body mass with a 1-min resting period between
sets. The last 3 repetitions of each set were performed explosively.
During the familiarization sessions, the subjects were instructed on
how to sit on the leg press, perform repetitions with a proper tech-
nique, and breathe while performing the exercise (expiration during
the concentric phase and inspiration during the eccentric phase to
avoid the Valsalva maneuver). The familiarization phase also served
to identify and record the correct position of each subject on the leg
press machine, to reproduce the same range of movement (ROM)
across the subjects and testing sessions (from 100º and 90º at the
hip and knee joints, respectively, to 180º or full extension).
During the F-V and muscle power testing procedure, the sub-
jects performed 2 sets of 1 repetition with increasing loads (10-kg
increments) from 40 % of their body mass. When the subjects failed
to lift a certain load, it was decreased by 5 kg until the one repeti-
tion maximum (1RM) was achieved. Force and velocity data during
the concentric phase of each repetition were recorded by a linear
position transducer device (T-Force System, Ergotech, Murcia,
Spain). The duration of the recovery time between sets was de-
signed according to the mean velocity exerted by the subjects in
the preceding repetition ( > 0.50 m · s
− 1
: 60 s of recovery time;
0.30–0.50 m · s
− 1
: 90 s of recovery time; < 0.30 m · s
− 1
: 120 s of re-
covery time). The subjects were continually encouraged to perform
each repetition as fast and strongly as possible. To ensure that all
the repetitions were performed at maximal speed, force and veloc-
ity data from each repetition were computed in a Microsoft Excel®
template (supplemental material), and the highest mean velocity
for each load was plotted. Because the F-V relationship was expect-
ed to follow a linear relationship [23], a linear regression equation
was tted simultaneously during the F-V evaluation. When the
highest mean velocity exerted against a certain load deviated more
than 0.03 m · s
− 1
from the estimated value based on the regression
analysis obtained with the preceding repetitions, a third repetition
was performed with that specic load. When a participant was not
able to exert his/her potentially maximal speed with a certain load,
and a deviation greater than 0.03 m · s
− 1
remained, that load was
discarded and the F-V relationship was computed considering the
remaining loads. The cut-o point of 0.03 m/s was selected based
on other studies implementing a velocity-based strength training
program [30] and according to pilot testing conducted in our lab-
oratory. In addition, after the 1RM determination and 5 min of re-
▶Table 1 Main characteristics of the subjects.
Variable Mean (SD) Range
Age (years) 75.8 (4.7) 70.2–84.9
BMI (kg/m
2
) 30.2 (4.4) 19.5–43.0
SPPB score 10.6 (2.1) 4.0–12.0
MMSE score 26.2 (3.0) 20.0–30.0
BMI: body mass index. MMSE: mini-mental state examination.
SD: standard deviation. SPPB: short physical performance battery
Downloaded by: National University of Singapore. Copyrighted material.
Alcazar J et al. The Force-Velocity Relationship in … Int J Sports Med
covery time, the subjects were asked to perform 2 sets of 3 repeti-
tions at 60 % 1RM in order to compare a single-repetition per set
(SR; a purely concentric muscle action) vs. a multiple-repetition per
set (MR; containing eccentric-concentric muscle actions) protocol.
The highest mean velocity produced from the rst repetition of
each set (only concentric) was compared against the highest mean
velocity produced from the second and third repetitions of each
set (eccentric-concentric).
Adverse events
Adverse events were carefully monitored for 3 weeks in which data
collection was conducted for each subject. An adverse event was
dened as any unfavorable or unintended event (pain, discomfort,
injury or accident) that occurred during the course of the study and
that might not necessarily be caused by the study procedures. In
case of an adverse event, it would be additionally classied as study-
related or non-study-related based on its origin and etiology.
Data analysis
Mean force and velocity values from each repetition, and force and
velocity values exerted at peak power within each repetition were ac-
quired from all the evaluations to compare both sets of data. Hereaf-
ter, mean data were used for the comparison between procedures.
As a reference method, the F-V relationship was calculated from
mean force and velocity data recorded from all the measured loads
(multiple-load (M-L) protocol), with 2 attempts per load, and an
additional attempt as mentioned in a previous section (2
+ 1
-at-
tempt protocol).
For comparison, we also calculated the F-V relationship not con-
sidering the third attempt (2-attempt protocol) if it was performed.
In addition, a short version of the entire protocol considering 3
loads (3-L) was also studied by selecting the rst, the middle and
the last load performed by the subjects.
In all cases, several variables were extracted from the F-V regres-
sion equation. Force at zero velocity (i. e., theoretical maximal isomet-
ric force) was obtained from the force-intercept (F
0
), while velocity at
zero force (i. e., maximal velocity with no load) was calculated as the
velocity-intercept (V
0
). The slope of the F-V relationship (S
FV
) was ob-
tained from the following equation (equation 1):
S
F
V
FV
=−
0
0
The linear relationship between force and velocity allowed for
maximal muscle power (P
max
) calculation using the following for-
mula (equation 2):
P
FV
max
=
×
00
4
The optimal load (L
opt
) at which the subjects exerted their P
max
was also calculated.
Statistical analysis
Sample size calculation was conducted based on the intra-class cor-
relation coecient (ICC) using the R package (R version 3.2.4 re-
vised) [10]. The arguments introduced were the preliminary ICC of
Pmax values (model 2,1) obtained in a group of 10 older subjects
(ICC = 0.977), the null hypothesis (ICC = 0.90), the number of rat-
ings of each subject (2), the number of tails (2) and the desired sta-
tistical power (0.80). A minimum of 15 older subjects was required
to satisfy the arguments introduced into the R package.
All data were examined statistically for normality of distribution
with the Shapiro-Wilk’s test, and standard descriptive statistics
were used for continuous variables.
Signicant dierences between protocols (collecting mean val-
ues vs. peak values; a 2-attempt vs. a 2
+ 1
-attempt protocol; an M-L
vs. a 3-L protocol; and an SR vs. an MR protocol) were assessed with
Student’s t-tests for independent samples.
Reliability of the F-V relationship and muscle power values ob-
tained from the dierent procedures were evaluated using dier-
ent approaches [20]. Within-subject variation was assessed by
means of the standard error of measurement (SEM), also reported
in relative terms (SEM %), and by the coecient of variation (CV);
changes in the mean of the dierent protocols between sessions
were analyzed with Student’s t-tests for dependent samples; and
retest correlation was evaluated using the ICC (model 2,1).
In addition, the dierent procedures were also compared re-
garding their associations with the SPPB score, as a measure of con-
struct validity. For this analysis, P
max
values were extracted from
the second evaluation, because of the greater experience gained
by the subjects after the rst evaluation. Bivariate linear and quad-
ratic regression analyses were performed because the relationship
between muscle power and physical function has been reported
to follow a quadratic relationship in some cases [2]. The coecient
of determination (R
2
) change was used to compare the linear and
quadratic models. Statistical analyses were performed with SPSS
v20 (SPSS Inc., Chicago, Illinois, USA), and the level of signicance
was set at α = 0.05.
Results
Testing procedure and adverse events
Force and velocity data showed a signicant linear relationship in
all subjects, with individual R
2
values ranging from 0.95 to 1.00,
whereas quadratic models did not signicantly improve the R
2
val-
ues (p > 0.05). The F-V relationship was evaluated by measuring
force and velocity data from 6.2 ± 1.8 loads and 13.2 ± 3.7 repeti-
tions per subject. On average, 0.8 ± 0.8 loads were discarded be-
cause they deviated from the F-V regression equation
( > 0.03 m · s
− 1
). Finally, 5.1 ± 1.9 loads were considered to obtain
the F-V relationship of the participants. The evaluation of the F-V
relationship was carried out in 26.9 ± 8.0 min per subject. No ad-
verse events were registered throughout the study.
Reliability of mean vs. peak values
There were no signicant dierences for any protocol between test-
ing sessions (▶Table 2). However, P
max
derived from peak values
was signicantly higher than that observed from mean values. Ab-
solute reliability was higher when mean values were considered in
comparison with peak values (▶Table 3). SEM % values of V
0
and
P
max
were signicantly lower and the ICC for P
max
signicantly high-
er when collecting mean compared with peak values.
Downloaded by: National University of Singapore. Copyrighted material.
Alcazar J et al. The Force-Velocity Relationship in … Int J Sports Med
Training & Testing
Thieme
Reliability of an SR vs. an MR protocol
There were no signicant dierences in mean muscle power exert-
ed at 60 % 1RM between testing sessions or between dierent pro-
tocols (▶Table 2). There were no signicant dierences in SEM %,
CV or ICC values between protocols (▶Table 4).
Reliability of protocols diering in the number of
attempts per load
The 2
+ 1
-attempt protocol showed no signicant dierences be-
tween testing sessions (▶Table 2). In contrast, P
max
measured by
the 2-attempt protocol was found to be signicantly dierent in
the second evaluation with respect to the rst evaluation. Both pro-
tocols were signicantly dierent in the rst evaluation. Reliability
was signicantly higher for the 2
+ 1
- than for the 2-attempt proto-
col regarding SEM % values for V
0
and P
max
, and ICC of V
0
values
(▶Table 5).
▶Table 2 Comparison between dierent protocols and testing sessions to assess the force-velocity relationship and muscle power.
Variable Procedure
Session 1 Session 2 Session 1 vs. 2
Mean (SD) Mean (SD) p value
P
max
(W) Data collection
Mean values 242.7 (93.0) * 246.5 (89.8) * 0.123
Peak values 505.5 (207.9) 530.2 (210.7) 0.064
Attempts per load
2
+ 1
-attempt 242.7 (93.0) * 246.5 (89.8) 0.123
2-attempt 235.3 (92.4) 244.7 (91.7) 0.024
¥
Number of loads
M-L 242.7 (93.0) 246.5 (89.8) 0.123
3-L 248.7 (97.7) 244.0 (87.6) 0.325
P
60 %1RM
(W) Repetitions per set
SR 223.6 (84.6) 227.9 (88.1) 0.562
MR 253.9 (90.5) 252.8 (93.1) 0.863
MR: multiple-repetition per set protocol. M-L: multiple-load protocol. P
max
: maximal muscle power. P
60 %1RM
: muscle power at 60 % 1RM. SD: standard
deviation. SR: single-repetition per set protocol. 2-attempt: 2 attempts per load. 2
+ 1
-attempt: 2 attempts per load plus an additional load when
indicated. 3-L: 3-load protocol. * Signicant dierences between protocols (p < 0.01).
¥
Signicant dierences between testing sessions (p < 0.05)
▶Table 3 Reliability comparison of force-velocity parameters obtained from mean and peak values.
Variable F
0
(N) V
0
(m/s) S
FV
P
max
(W) L
opt
(kg)
SEM [95 % CI]
Mean values 99.4 [88.6–114.3] 0.05 [0.05-0.05] 268.5 [231.7–319.2] 8.0 [7.0–9.6] 5.1 [4.5–5.8]
Peak values 123.7 [110.3–140.7] 0.26 [0.23–0.29] 176.1 [151.2–210.8] 44.2 [37.8–53.1] 6.3 [5.6–7.2]
SEM [95 % CI] ( %)
Mean values 9.2 [8.1–10.5] 5.7 * [5.2–6.3] 21.1 [18.2–25.1] 3.3 * [2.8–3.9] 9.2 [8.1–10.5]
Peak values 11.3 [10.1–12.8] 13.5 [12.2–15.3] 28.3 [24.3–33.8] 8.5 [10.1–12.8] 11.3 [10.1–12.8]
CV [95 % CI] ( %)
Mean values 5.6 [2.5–8.7] 4.8 [2.2–7.5] 10.1 [4.6–15.7] 2.6 [1.5–3.7] 5.5 [2.5–8.7]
Peak values 7.6 [3.7–11.5] 10.6 [6.8–14.4] 17.7 [10.8–24.5] 5.8 [3.5–8.2] 7.6 [3.7–11.5]
ICC [95 % CI]
Mean values 0.91 [0.81–0.96] 0.94 [0.86–0.97] 0.73 [0.47–0.87] 0.99 * [0.98–1.00] 0.91 [0.81–0.96]
Peak values 0.86 [0.70–0.94] 0.77 [0.54–0.90] 0.59 [0.25–0.80] 0.96 [0.90–0.98] 0.86 [0.70–0.94]
CI: condence interval. CV: coecient of variation. F
0
: force-intercept. ICC: intra-class correlation coecient. L
opt
: optimal load. P
max
: maximal power.
SEM: standard error of measurement. S
FV
: slope of the force-velocity relationship. V
0
: velocity-intercept. * Signicant dierences between mean and
peak values (p < 0.05)
▶Table 4 Reliability comparison of mean muscle power exerted at 60 %
1RM between the single- and the multiple-repetition per set protocols.
Variable SR MR
SEM [95 % CI] (W) 23.9 [20.5–28.6] 20.7 [17.8–24.5]
SEM [95 % CI] ( %) 10.6 [9.1–12.7] 8.2 [7.0–9.7]
CV [95 % CI] ( %) 8.2 [5.3–11.2] 6.5 [4.4–8.5]
ICC [95 % CI] 0.93 [0.83–0.97] 0.95 [0.89–0.98]
CI: condence interval. CV: coecient of variation. ICC: intra-class
correlation coecient. MR: multiple-repetition per set protocol. SEM:
standard error of measurement. SR: single-repetition per set
protocol. * Signicant dierences between the SR and MR protocols
(p < 0.05)
Downloaded by: National University of Singapore. Copyrighted material.
Alcazar J et al. The Force-Velocity Relationship in … Int J Sports Med
Reliability of protocols diering in the number
of loads
There were no signicant dierences in P
max
between testing ses-
sions or protocols regarding the number of loads considered
(▶Table 2). However, the M-L protocol was signicantly more reli-
able than the 3-L protocol, considering their SEM % values for F
0
,
V
0
, P
max
and L
opt
(▶Table 5).
Validity of the dierent protocols regarding physical
function
Collecting mean values was found to be superior (curvilinear rela-
tionship; R
2
= 0.34) than peak values (linear relationship; R
2
= 0.15)
to predict the participants’ SPPB scores. Both the SR and MR pro-
tocols showed a signicant and similar linear relationship with phys-
ical function (R
2
= 0.19 and 0.18, respectively). R
2
values were also
similar between the 2
+ 1
- and 2-attempt protocols (both curvilin-
ear relationships; R
2
= 0.34 and 0.33, respectively), and between
the M-L and 3-L protocols (both curvilinear relationships; both
R
2
= 0.34).
Discussion
Our ndings showed that a protocol collecting mean force and ve-
locity values from either concentric or eccentric-concentric mus-
cle actions performed against multiple progressive loads was an
optimal strategy to evaluate the F-V relationship and muscle power
in older adults in terms of reliability and external validity, explain-
ing up to 34 % of the variability in physical function.
The great heterogeneity reported in the literature investigating
muscle power in older adults is likely to be partially explained by
the dierent methods and protocols used [7, 37]. P
max
values sig-
nicantly diered when collecting mean or peak values. In young
adults, there are inconclusive results regarding reliability of mus-
cle power testing, with some evidence in favor of using mean val-
ues [12, 14], and other evidence supporting the use of peak values
[13]. In older adults, one previous study compared the utility of
both approaches, and mean values were more strongly associated
with the older subjects’ muscle quality and size [38]. In this line,
we found that mean values were signicantly more reliable and
better associated with physical function in comparison with peak
values, thus mean values might be preferred when evaluating older
individuals.
In addition, muscle power can be measured using purely con-
centric muscle contractions (SR protocol) or stretch-shortening
cycle (SSC) actions where eccentric and concentric contractions
are performed (MR protocol). Though the SSC is believed to in-
crease neuromuscular performance in young adults [26], contra-
dictory results exist regarding older adults [5, 22]. In our study, the
SSC did not improve power values at 60 % 1RM over only-concen-
tric repetitions, and reliability was similar between both protocols.
In young adults, an SSC protocol was found to have a greater error
than a solely-concentric protocol, although both of them were
highly reliable [29]. In the absence of signicant dierences be-
tween the two protocols, we prefer to use the SR protocol due to
its simplicity, and overall because it might be a safer alternative for
older adults compared with MR protocols, which impose a higher
mechanical and cardiovascular load [27], leading to an increased
risk of injury or cardiovascular event. However, future studies
should evaluate the inuence of the SSC on muscle power of older
adults over the whole F-V relationship.
The main novelty that this study introduces is the implementa-
tion of a systematic procedure by which force and velocity data can
▶Table 5 Reliability comparison of force-velocity parameters obtained from protocols diering in the number of attempts per load and/or number of
loads.
Variable F
0
(N) V
0
(m/s) S
FV
P
max
(W) L
opt
(kg)
SEM [95 % CI]
M-L [2
+ 1
-attempt] 99.4 [88.6–114.3] 0.05 [0.05-0.05] 268.5 [231.7–319.2] 8.0 [7.0–9.6] 5.1 [4.5–5.8]
M-L [2–attempt] 98.3 [90.1–108.1] 0.08 [0.07–0.08] 271.5 [241.9–309.3] 13.4 [12.1–15.1] 5.0 [4.6–5.5]
3-L [2
+ 1
-attempt] 142.0 [129.8–156.5] 0.09 [0.09–0.10] 278.1 [247.5–317.1] 16.3 [14.7–18.3] 7.2 [6.6–8.0]
SEM [95 % CI] ( %)
M-L [2
+ 1
-attempt] 9.2 [8.1–10.5] 5.7 [5.2–6.3] 21.1 [18.2–25.1] 3.3 [2.8–3.9] 9.2 [8.1–10.5]
M-L [2–attempt] 9.1 [8.4–10.0] 8.5 * [8.0–9.1] 21.3 [19.0–24.3] 5.6 * [5.0–6.3] 9.1 [8.4–10.0]
3-L [2
+ 1
-attempt] 12.9 * [11.8–14.3] 10.2 * [9.6–10.9] 21.6 [19.3–24.7] 6.6 * [6.0–7.4] 12.9 * [11.8–14.3]
CV [95 % CI] ( %)
M-L [2
+ 1
-attempt] 5.6 [2.5–8.7] 4.8 [2.2–7.5] 10.1 [4.6–15.7] 2.6 [1.5–3.7] 5.5 [2.5–8.7]
M-L [2–attempt] 6.3 [3.3–9.3] 6.1 [2.8–9.4] 12.0 [6.5–17.5] 4.4 [2.0–6.9] 6.3 [3.3–9.3]
3-L [2
+ 1
-attempt] 9.9 [6.6–13.1] 8.5 [5.8–11.2] 17.8 [12.2–23.5] 4.3 [2.7–5.8] 9.9 [6.6–13.1]
ICC [95 % CI]
M-L [2
+ 1
-attempt] 0.91 [0.81–0.96] 0.94 [0.86–0.97] 0.73 [0.47–0.87] 0.99 [0.98–1.00] 0.91 [0.81–0.96]
M-L [2–attempt] 0.92 [0.82–0.96] 0.85 * [0.68–0.81] 0.75 [0.51–0.88] 0.98 [0.93–0.99] 0.92 [0.82–0.96]
3-L [2
+ 1
-attempt] 0.84 [0.67–0.93] 0.81 [0.60–0.91] 0.75 [0.50–0.88] 0.97 [0.93–0.99] 0.84 [0.67–0.93]
CI: condence interval. CV: coecient of variation. F
0
: force-intercept. ICC: intra-class correlation coecient. L
opt
: optimal load. M-L: multiple-load
protocol. P
max
: maximal power. SEM: standard error of measurement. S
FV
: slope of the force-velocity relationship. V
0
: velocity-intercept. 2-attempt: 2
attempts per load. 2
+ 1
-attempt: 2 attempts per load plus an additional load when indicated. 3-L: 3-load protocol. * Signicantly dierent compared
with the M-L [2
+ 1
-attempt] protocol (p < 0.05)
Downloaded by: National University of Singapore. Copyrighted material.