Sayers 2010

A COMPARISON OF HIGH-SPEED POWER TRAINING
AND TRADITIONAL SLOW-SPEED RESISTANCE

TRAINING IN OLDER MEN AND WOMEN
STEPHEN P. SAYERS AND KYLE GIBSON
Neuromuscular Research and Rehabilitation Laboratory, Department of Physical Therapy, University of Missouri,
Columbia, Missouri
ABSTRACT INTRODUCTION
A
Sayers, SP and Gibson, K. A comparison of high-speed power n inevitable consequence of aging is loss of muscle
training and traditional slow-speed resistance training in older mass and concomitant reductions in muscle
men and women. J Strength Cond Res 24(12): 3369–3380, strength and functional capacity (22). One way
2010—Muscle power, the product of force 3 velocity, is to increase muscle strength is through resistance
a critical determinant of function in older adults. Resistance
training (RT), and numerous studies have described the
success of this method in older adults (10,13,24,28,32,35,37).
training (RT) at high speed has been shown to improve peak
In the past 2 decades, however, muscle power has emerged as
muscle power in this population; however, different functional
a critical muscle performance characteristic in this popula-
tasks may benefit from the improvement of power at values
tion (3,5,9,12,36). Muscle power is the product of force 3
other than ‘‘peak’’ values, for example, tasks that require velocity and reflects the ability of the muscle to produce force
a greater velocity component or a greater force component. rapidly (3). Decreases in muscle mass (22) and changes in the
This study compared the effect of high-speed RT on muscle ability of the muscle to contract at high speed (such as type II
performance (peak power [PP] and its components [PP force muscle fiber loss and atrophy [19,20]; slowing of contractile
and PP velocity] and overall peak velocity [VEL]) across a broad velocity [17,18]; and slowing of nerve conduction velocity
range of external resistances. Thirty-eight older men and women [29,38]) may accelerate the age-related decreases in power
were randomized to high-speed power training at 40% of the compared to strength. As a result, muscle power begins to
1-repetition maximum (1RM) (n = 13 [74.1 6 6.4 years]); decline sooner and more rapidly than muscle strength with
traditional RT at 80% 1RM (n = 13 [70.1 6 7.0 years]); or aging (25) and at up to twice the rate than strength (3–4 vs.
control (n = 12 [72.8 6 4.1 years]). Measures of muscle 1–2%) after age 60 (31).
performance were obtained at baseline and after the 12-week A number of studies in the older adult literature have
demonstrated that muscle power is a more critical variable to
training intervention. Muscle power and 1RM strength improved
functional decline in older adults than muscle strength
similarly with both high-speed and traditional slow-speed RT.
(3,5,12,36). Other studies have demonstrated that high-speed
However, speed-related muscle performance characteristics,
power training in older adults may also improve muscle
PP velocity and overall VEL, were most positively impacted by
power (11,15,23,27) or functional performance (26) com-
high-speed power training, especially at lower external pared to traditional RT with a strengthening component.
resistances. Because gains in speed-related measures with A key outcome variable in these studies has been peak power
high-speed training compared to traditional RT do not come (PP). This outcome measure has traditionally been
at the expense of other muscle performance outcomes, we determined by measuring power output at several percen-
recommend using an RT protocol in older adults that tages of the 1-repetition maximum (1RM) (3,6–8,11,30,33),
emphasizes high-speed movements at low external resistances. and choosing the highest power output value across that
range of external resistances. Power typically increases as
KEY WORDS muscle power, velocity, resistance training, aging external resistance is increased (see Figure 1) and decreases at
the highest external resistances (11,33). The PP has typically
been found at or near 70% 1RM in this population (11).
The PP, however, is not the only muscle power outcome
Address correspondence to Stephen P. Sayers, [email protected]. variable that is valuable as a contributor to muscle performance.
24(12)/3369–3380 We reported previously that muscle power developed at high
Journal of Strength and Conditioning Research velocity (40% 1RM) and the corresponding contraction
Ó 2010 National Strength and Conditioning Association velocity at which that power was developed were better
VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3369
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
High-Speed Power Training in Older Adults
keep from falling vs. slowly getting up from a chair). Few

studies in older adults have focused on the components of
muscle power (such as force at PP or velocity at peak power
[PPV]) and how these variables are impacted by RT.
We believe studies in this area have been limited by little or
no examination of the following: (a) muscle power at values
other than the ‘‘peak’’ value across a range of external
resistances and (b) changes in muscle performance (power
and its components [force and velocity] and overall peak
velocity [VEL]) with RT. In addition, few studies have
evaluated muscle performance variables relative to both the
posttraining 1RM and baseline 1RM (which may be more
applicable to real world situations where external resistances
are fixed and not variable). The purpose of the study was to
Figure 1. Data representing a power curve at baseline for the 38 older determine whether low-load, high-speed power training and
men and women in the study. Solid squares (n) represent peak muscle traditional high-load, slow-speed RT may have differing

power and solid circles ( ) represent peak velocity obtained across effects on power output obtained across a range of external
a range of external resistance (40–90% of the 1-repetition maximum
[1RM]). resistances. In addition, we sought to explore the impact of
these RT regimens on the determinants of muscle power, that
is, velocity and force at PP, and VEL; variables that could be
critical to functioning but not commonly reported. We
indicators of functional declines in older adults than muscle hypothesized that high-speed power training would improve
strength (1RM) or muscle power at slower velocities (70% muscle power to a greater extent than slow-speed RT through
1RM) (34). These findings suggest that even if power output its effects on speed-related muscle performance character-
is similar at certain points along a power curve, the different istics (PPV and overall VEL).
velocities at which power is obtained could be a key factor in
functional responses. For example, Figure 1 shows similar
muscle power outputs obtained at both 40% 1RM and 90%
1RM; however, the velocity at which these power outputs METHODS
was developed was very different, with power developed at Experimental Approach to the Problem
40% 1RM requiring a greater velocity component and power The goal of this study was to compare 12 weeks of explosive
at 90% 1RM requiring a greater force component. Certain high-speed power training vs. traditional slow-speed RT with
functional tasks may require power with a greater velocity a strengthening component in older men and women and
component, whereas others may require power with a greater examine the effects of this training on muscle performance
force component. This makes intuitive sense when one characteristics. Dependent variables in the study consisted of
considers the different functional tasks that older adults PP and its components (PPVand force at PP) and VEL across
encounter each day (e.g., moving the lower limb quickly to a range of external resistances (40–90% of the 1RM).
TABLE 1. Subject characteristics.*†
HSPT (n = 13: 6m, 7f) STR (n = 13: 5m, 8f) CON (n = 12: 3m, 9f) p
Age 74.1 6 6.4 70.1 6 7.0 72.8 6 4.1 0.38

Height (cm) 170.3 6 9.8 169.5 6 10.6 165.8 6 9.5 0.49
Weight (kg) 76.6 6 18.5 71.0 6 25.4 78.6 6 17.7 0.66
GDS (0–30) 4.2 6 5.0 6.7 6 4.8 5.3 6 3.9 0.40
MMSE (0–30) 28.3 6 1.6 28.2 6 2.7 28.6 6 1.0 0.85
Medications (no.) 5.6 6 2.6 4.5 6 3.1 6.0 6 3.2 0.45
Falls in the past year (no.) 0.15 6 0.4 0.54 6 0.8 0.17 6 0.6 0.20
*HSPT = high-speed power training group; STR = slow-speed resistance training group; CON = control group; GDS = geriatric
Depression Scale; MMSE = mini-mental state examination.
†Data are mean 6 SD.
the TM
3370 Journal of Strength and Conditioning Research
TABLE 2. Muscle performance variables across a range of external resistances at baseline and after 12 weeks of resistance training (using loads relative to
posttraining 1RM).*
Variable 40% 1RM 50% 1RM 60% 1RM 70% 1RM 80% 1RM 90% 1RM
Peak power (W) HSPT Pre: 298.5 6 136.2 Pre: 321.8 6 133.2 Pre: 329.4 6 137.9 Pre: 317.4 6 139.7 Pre: 293.8 6 136.1 Pre: 250.8 6 140.3
Post: 378.3 6 145.9 Post: 416.1 6 164.3 Post: 415.1 6 161.2 Post: 380.0 6 153.6 Post: 340.8 6 140.6 Post: 273.7 6 120.1
% Change: 27 % Change: 29 % Change: 26 % Change: 20 % Change: 16 % Change: 9
STR Pre: 324.6 6 154.3 Pre: 360.7 6 183.5 Pre: 364.4 6 184.1 Pre: 357.2 6 189.0 Pre: 325.8 6 175.3 Pre: 306.8 6 157.1
CON Pre: 270.3 6 137.1 Pre: 294.8 6 148.4 Pre: 294.4 6 145.8 Pre: 285.9 6 140.6 Pre: 265.4 6 126.0 Pre: 244.1 6 120.6
Peak power HSPT Pre: 69.4 6 18.2 Pre: 84.7 6 22.0 Pre: 99.0 6 26.7 Pre: 113.4 6 31.4 Pre: 127.1 6 36.3 Pre: 139.5 6 40.6
torque (Nm) Post: 84.2 6 23.6 Post: 102.8 6 30.2 Post: 120.9 6 36.7 Post: 138.9 6 42.5 Post: 154.1 6 50.6 Post: 168.6 6 52.7
velocity (Nms21) Post: 4.4 6 0.8 Post: 4.0 6 0.7 Post: 3.4 6 0.6 Post: 2.7 6 0.6 Post: 2.2 6 0.6 Post: 1.6 6 0.5
Journal of Strength and Conditioning Research

the
VOLUME 24 | NUMBER 12 | DECEMBER 2010 |

Peak velocity HSPT Pre: 4.3 6 0.9 Pre: 3.7 6 0.9 Pre: 3.3 6 0.8 Pre: 2.8 6 0.7 Pre: 2.3 6 0.7 Pre: 1.8 6 0.7
(Nms21) Post: 4.5 6 0.8 Post: 4.1 6 0.8 Post: 3.5 6 0.6 Post: 2.7 6 0.6 Post: 2.2 6 0.6 Post: 1.6 6 0.6
TM
*HSPT = high-speed power training group; STR = slow-speed resistance training group; CON = control group.
| www.nsca-jscr.org
3371
3372

TABLE 3. Muscle performance variables across a range of external resistances at baseline and after 12 weeks of resistance training (using loads relative to baseline 1RM).*
Variable 40% 1RM 50% 1RM 60% 1RM 70% 1RM 80% 1RM 90% 1RM
Journal of Strength and Conditioning Research
the
(W) Post: 383.3 6 113.3 Post: 426.6 6 133.6 Post: 456.6 6 136.6 Post: 456.3 6 150.2 Post: 457.6 6 149.3 Post: 431.3 6 162.4
torque Post: 73.4 6 15.8 Post: 89.6 6 21.6 Post: 104.6 6 25.9 Post: 121.5 6 31.2 Post: 135.9 6 36.9 Post: 152.8 6 41.9
(Nm) % Change: 5 % Change: 5 % Change: 5 % Change: 6 % Change: 6 % Change: 8
TM
velocity Post: 5.2 6 0.6 Post: 4.7 6 0.6 Post: 4.4 6 0.5 Post: 3.8 6 0.5 Post: 3.4 6 0.5 Post: 2.8 6 0.5
(Nms21) % Change: 18 % Change: 22 % Change: 27 % Change: 28 % Change: 40 % Change: 50
Peak velocity HSPT Pre: 4.5 6 0.9 Pre: 3.9 6 0.9 Pre: 3.5 6 0.8 Pre: 3.0 6 0.7 Pre: 2.5 6 0.7 Pre: 1.9 6 0.8
(Nms21) Post: 5.3 6 0.9 Post: 4.8 6 0.6 Post: 4.4 6 0.6 Post: 3.8 6 0.5 Post: 3.5 6 0.5 Post: 2.9 6 0.5
*HSPT = high-speed power training group; STR = slow-speed resistance training group; CON = control group.
the TM
Journal of Strength and Conditioning Research | www.nsca-jscr.org
study. Eleven dropped out of

the study before they had begun
their training: Eight could not
find time in their schedules to
participate, and 3 withdrew
after being informed of their
randomization into the control
group. Of the 41 remaining
participants who began the
exercise study, our retention
rate was excellent (93%). Only
3 participants withdrew during
the intervention, 1 in each of the
3 groups. One participant in the
HSPT group withdrew because
of a recurring back injury (not
study related), 1 in the STR
group withdrew because of pain
in her knee, and 1 in the CON
group could no longer find
transportation to and from the
facility. This left 38 older men
and women (HSPT: n = 13;
STR: n = 13; CON: n = 12)
remaining who completed the
study. Subject characteristics
are presented in Table 1. This
research project was approved
by the University of Missouri
(MU) Institutional Review
Board, and written informed
consent was obtained from all
participants.
Figure 2. Change in knee extension peak power across a range of external resistances (40–90% of the 1-repetition
maximum [1RM]) relative to A) posttraining 1RM and B) baseline 1RM for high-speed power training group (HSPT), Procedures
traditional slow-speed resistance training group (STR), and control (CON). All data represent mean 6 SEM. Maximal Strength and Power
Determination. Leg press (LP)
and seated knee extension (KE)
Subjects 1RM were obtained using Keiser a420 pneumatic (air
Fifty-two participants were originally enrolled in the study resistance) equipment (Fresno, CA, USA). Keiser’s a420
and randomized to 1 of 3 treatment arms: High-Speed Power computer-interfaced pneumatic RT equipment is designed to
Training at 40% 1RM (HSPT), Slow-Speed RT with capture the measures of strength and additional muscle
a strengthening component at 80% 1RM (STR), or Control performance measures calculated with computer software.
(CON). All participants were healthy by self-report and During the concentric actions as the exercise arm is moved
community-dwelling but possessed some limitations in either through its range of motion, a piston is driven into a cylinder
function or mobility. All participants were seen by a study where it encounters the mechanical resistance of the air
physician before they were allowed to take part in the study. pressure in the system. The a420 equipment captures
Exclusion criteria consisted of history of heart disease, measures of peak muscle power, PPV, force at PP, and
osteoarthritis, severe visual impairment, presence of neuro- VEL during the concentric portion of each contraction by
logical disease, pulmonary disease requiring the use of sampling the system pressure at 400 Hz and making
oxygen, uncontrolled hypertension, hip fracture, or lower calculations based on an appropriate algorithm. For brevity,
extremity joint replacement in the past 6 months, and current only the KE data from the study are presented here.
participation in a structured exercise. Fourteen of the One repetition maximum measures were obtained by
52 enrolled study subjects discontinued participation in the progressively increasing the resistance until the subject was
training period. All outcome

measures of 1RM and power
across a range of external
resistances (40–90% 1RM)
were performed twice during
the baseline period to establish
reliability and once at the end
of the 12-week period. A sec-
ond posttraining measure of
muscle performance (PP, PPV,
PPT, and VEL) across the
range of external resistances
(40–90% 1RM) was also
obtained 1 week after the
power test performed at 12
weeks. However, the loads used
were relative to the initial
baseline 1RM (not relative to
the higher 1RM resulting from
the training intervention). Be-
cause this second power test
was introduced after the study
had commenced, only 27 of the
38 participants underwent this
final evaluation (HSPT: n = 9;
STR: n = 11; CON: n = 7).
Resistance Training Protocol. Vol-

unteers randomized into HSPT
and STR groups exercised 3
times per week for 12 weeks
using computer-interfaced Ke-
iser a420 pneumatic LP and KE
RT equipment. For HSPT, each
Figure 3. Change in knee extension peak power velocity across a range of external resistances (40–90% of the training session consisted of 3
1-repetition maximum [1RM]) relative to A) posttraining 1RM and B) baseline 1RM for high-speed power training sets of 12–14 repetitions at 40%
group (HSPT), traditional slow-speed resistance training group (STR), and control (CON). All data represent
mean 6 SEM.
of the 1RM. Participants per-
formed each movement at high
velocity during the concentric
phase of each repetition (‘‘as
no longer able to push out 1 repetition successfully. The fast as possible’’), paused for 1 second, and performed the
ratings of perceived exertion scale (4) was used to assist in eccentric portion of the contraction over 2 seconds.
evaluating when 1RM (combined with perceived maximal Volunteers randomized into the STR group also exercised 3
effort) was successfully reached. Peak muscle power was times per week for 12 weeks using the same Keiser a420 RT
obtained at 40, 50, 60, 70, 80, and 90% of the 1RM equipment. For STR, each training session consisted of 3 sets
approximately 30 minutes after the 1RM measure was of 8–10 repetitions at 80% of the 1RM. Repetitions were
obtained (7,30). Participants were instructed to exert ‘‘as fast higher in the HSPT group to more closely equate work
as possible’’ at each of these relative percentages of the 1RM. performed between the training groups and to remain
Three attempts were made at each resistance, and the consistent with RT exercise guidelines (1). The participants
greatest PP output obtained at each resistance was used in the performed each movement at a slow velocity using proper
analysis. The PPV, the torque at peak power (PPT), and VEL technique (2 seconds for concentric phase of the repetition),
were obtained for each external resistance across the power paused for 1 second, and performed the eccentric portion of
curve. The 1RM was measured every 2 weeks in the HSPT the contraction over over a 2-second duration. Thus, the key
and STR groups only, and relative training intensity was differences in the RT protocols between HSPT and STR was
adjusted accordingly to ensure adequate overload during the the percentage of the 1RM at which each group trained (40 vs.
the TM
the TM
(group by time) repeated meas-

ures ANOVA was run to eval-
uate differences among groups
in KE 1RM strength from
baseline to posttesting. Change
scores in PP, PPV, PPT, and
VEL with training were ob-
tained by subtracting the base-
line values from the
posttraining value. To evaluate
differences among groups in the
change in muscle performance
across 6 external resistances
with training, a 3 3 6 (group
by condition) repeated-
measures ANOVA was run for
each muscle performance mea-
sure. When equal variances
were not observed using
Mauchley’s test of sphericity,
corrections to the F-ratio were
applied using Greenhouse–
Geyser Epsilon values. In sta-
tistically significant repeated
measures ANOVA models
(group main effects or group
by time interactions), post hoc
testing using Tukey’s HSD
(when main effects were pres-
ent) or independent samples
t-tests (when interactions were
present) was performed to
determine the source of statis-
tically significant differences.
Figure 4. Change in knee extension peak power torque across a range of external resistances (40–90% of the
1-repetition maximum [1RM]) relative to A) posttraining 1RM and B) baseline 1RM for high-speed power training
Statistical significance for
group (HSPT), traditional slow-speed resistance training group (STR), and control (CON). All data represent Chi-square tests (Mauchley),
mean 6 SEM. ANOVA models, and t-tests
was accepted at p # 0.05. Data
are reported as mean 6 SD.
80%) and the speed at which the weights were moved RESULTS
concentrically. The CON group met three times a week to
Baseline Measures
participate in the warm-up and stretching exercises but
There were no differences in age, height, weight, depressive
performed no RT exercise. The HSPT and STR groups
symptoms, cognitive function, number of medications, or falls
participated in the same warm-up and stretching exercises as
in the past year among groups at baseline (all p . 0.05; see
the CON group did before their training.
Table 1). Knee extension 1RM strength among groups
Statistical Analyses (HSPT:158 6 56 N m; STR: 173 6 66 N m; CON: 145 6 62
Descriptive statistics were run on all variables. One-way N m was not different at baseline (F = 0.62; p = 0.55).
analysis of variance (ANOVA) was run to evaluate baseline When comparing muscle performance characteristics of our 3
differences among groups in subject characteristics and KE groups at baseline (HSPT: n = 13; STR: n = 13; CON: n = 12)
1RM strength. To evaluate baseline differences among groups during performance of a power test across a range of external
in muscle performance (PP, PPV, PPT, and VEL) across resistances (40–90%1RM), we found no group main effects or
a range of external resistances, a 3 3 6 (group by condition group by condition interactions in PP (Group: F = 0.60; p =
[i.e., 40, 50, 60, 70, 80, 90% 1RM]) repeated measures ANOVA 0.56; G 3 C: F = 0.57; p = 0.84), PPV (Group: F = 1.8; p = 0.18;
was run for each muscle performance measure. A 3 3 2 G 3 C: F = 0.56; p = 0.85), PPT (Group: F = 0.72; p = 0.49; G 3
performance characteristics at
baseline were significant (all
p , 0.001) as expected.
Changes from Baseline

Differences among groups in
KE 1RM strength were com-
pared from baseline to post-
training. There was a significant
time main effect, with KE 1RM
strength increasing 19% over 12
weeks of training (F = 27.9; p ,
0.001). There was no significant
group main effect (F = 1.8; p =
0.18), but there was a significant
group 3 time interaction (F =
6.9; p = 0.003) suggesting there
was a difference in the pattern
of response to the different
training methods. The KE
1RM increased in the HSPT
group (25%) compared to in
the STR (26%) and CON (0%)
groups.
The changes in PP, PPV,
PPT, and VEL across a range
of external resistances (40–90%
1RM) were compared from
baseline to posttraining. The
loads used to assess posttraining
PP, PPV, PPT, and VEL were
relative to (a) the final post-
training 1RM and (b) the base-
line 1RM. Sample sizes for the
Figure 5. Change in knee extension peak velocity across a range of external resistances (40–90% of the
1-repetition maximum [1RM]) relative to A) posttraining 1RM and B) baseline 1RM for high-speed power training latter evaluations (b) were as
group (HSPT), traditional slow-speed resistance training group (STR), and control (CON). All data represent follows: HSPT (n = 9), STR
mean 6 SEM.
(n = 11), CON (n = 7). The
values for all of our baseline and
posttraining PP, PPV, PPT, and
C: F = 1.1; p = 0.38), or VEL (Group: F = 1.7; p = 0.19; G 3 C: VEL for all participants and the subset of participants can be
F = 0.51; p = 0.88) among groups. Condition main effects for all seen in Tables 2 and 3, respectively.
muscle performance characteristics at baseline were all
significant (all p , 0.001) as expected because of the parabolic Peak Power. In the first analysis (a), repeated-measures
nature of the power curve (Figure 1) and the decrease in ANOVA using the change scores relative to the posttraining
velocity and increase in force typically observed across this 1RM indicated a significant group main effect (F = 12.5; p ,
range of increasing resistances (40–90% 1RM) (33). 0.001) and condition main effect (F = 6.4; p , 0.001) but no
Comparing the baseline muscle performance characteristics in significant group by condition interaction (F = 0.79; p = 0.64)
a subset of participants in our 3 groups (HSPT: n = 9; STR: (see Figure 2A). To further evaluate the significant group
n = 11; CON: n = 7) during performance of a power test across main effect, post hoc testing revealed that both HSPT (p ,
a range of external resistances (40–90%1RM), we also found 0.001) and STR (p = 0.001) demonstrated greater PP output
no differences in baseline PP (Group: F = 0.23; p = 0.80; G 3 C: compared to CON but were not significantly different from
F = 0.41; p = 0.94), PPV (Group: F = 1.3; p = 0.30; G 3 C: F = each other (p = 0.68). In the second analysis (b), repeated--
0.40; p = 0.95), PPT (Group: F = 0.12; p = 0.89; G 3 C: F = 0.30; measures ANOVA using change scores relative to the
p = 0.98), or VEL (Group: F = 1.2; p = 0.32; G 3 C: F = 0.36; baseline 1RM indicated a significant group main effect
p = 0.96) among groups. Condition main effects for all muscle (F = 9.5; p = 0.001), condition main effect (F = 13.6; p , 0.001)
the TM
the TM
and group by condition interaction (F = 2.7; p = 0.005) (p = 0.02) only but showed a strong trend at 60% 1RM
(Figure 2B). To further evaluate the significant interaction, post (p = 0.08). The HSPT group demonstrated greater VEL
hoc testing revealed that both HSPT and STR demonstrated compared to the STR group at 50% 1RM (p = 0.005) and
greater PP output compared to CON at 40% 1RM (both: 60% 1RM (p = 0.03) and showed a strong trend at 40% 1RM
p # 0.05), 50% 1RM (both: p # 0.02), 60% 1RM (both: p # (p = 0.06). The HSPTand STR values were not different from
0.03), 70% 1RM (both: p # 0.001), 80% 1RM (both: p # 0.002), the CON value at 70% 1RM (both: p . 0.05), 80% 1RM
and 90% 1RM (both: p # 0.004) but were not significantly (both: p . 0.05), or 90% 1RM (both: p . 0.05). In the second
different from each other at any time point (all: p . 0.05). analysis (b), repeated-measures ANOVA using the change
scores relative to the baseline 1RM indicated a significant
Velocity at Peak Power. In the first analysis (a), repeated-meas- group main effect (F = 11; p , 0.001) and condition main
ures ANOVA using the change scores relative to the effect (F = 3.2; p = 0.009) but no group by condition
posttraining 1RM indicated a significant group main effect interaction (F = 0.81; p = 0.62) (Figure 5B). To further evalu-
(F = 4.2; p = 0.05) and condition main effect (F = 2.5; p = ate the significant group main effect, post hoc testing revealed
0.03) but no significant group by time interaction (F = 1.4; p = that HSPTdemonstrated greater VEL compared to STR (p =
0.17) (Figure 3A). To further evaluate the significant group 0.03) and CON (p , 0.001). The STR group demonstrated
main effect, post hoc testing revealed that HSPT demon- greater VEL compared to the CON group (p = 0.01).
strated greater PPV compared to STR (p = 0.03) but not
CON (p = 0.23). The STR and CON values were not
significantly different from each other (p = 0.30). In the DISCUSSION
second analysis (b), repeated-measures ANOVA using the Over the past 2 decades, muscle power has been examined
change scores relative to the baseline 1RM indicated extensively in the older adult population, both as a predictor
a significant group main effect (F = 11.8; p , 0.001), of performance (3,5,34,36) and as an outcome variable in RT
condition main effect (F = 3.3; p = 0.007) but no group by and power training studies (6–8,11,14–16,27). Previous
condition interaction (F = 1.1; p = 0.40) (Figure 3B). To studies on this topic have been limited by little or no
further evaluate the significant group main effect, post hoc examination of (a) muscle power at values other than the
testing revealed that HSPT demonstrated greater PPV ‘‘peak’’ value across the full spectrum of the power curve and
compared to STR (p = 0.03) and CON (p , 0.001). The (b) changes in the different components of muscle power
STR group did not demonstrate greater PPVcompared to the (force at PP and PPV) with RT. The present study addressed
CON group (p = 0.06) but showed a strong trend. these 2 limitations by examining how 2 different RT
Torque (Force) at Peak Power. In the first analysis (a), protocols, 1 approach emphasizing high-speed movements
repeated-measures ANOVA using the change scores relative at low resistance (high-speed power training) and the other
to the posttraining 1RM indicated a significant group main approach emphasizing slow-speed movements at high
effect (F = 6.5; p = 0.004), condition main effect (F = 23; p , resistance (traditional RT with a strengthening component),
0.001), and group by condition interaction (F = 3.7; p , 0.001) contribute to muscle performance (i.e., muscle power and its
(Figure 4A). To further evaluate the significant interaction, post components [force at PP and PPV] and VEL) across a range
hoc testing revealed that both HSPT and STR demonstrated of external resistances. In addition to focusing on changes in
greater PPTcompared to CON at 40% 1RM (both: p # 0.003), muscle performance relative to the posttraining 1RM, we
50% 1RM (both: p # 0.003), 60% 1RM (both: p # 0.009), 70% also included an evaluation of changes in muscle perform-
1RM (both: p # 0.004), 80% 1RM (both: p # 0.02), and 90% ances relative to the baseline 1RM, which may be more
1RM (both: p # 0.03) but were not significantly different from applicable to real-world situations where external resistances
each other at any time point (all: p . 0.05). In the second are fixed and not variable. This investigation is one of the first
analysis (b), repeated-measures ANOVA using the change to focus on how high-speed RT impacts muscle performance
scores relative to the baseline 1RM indicated a significant main across a range of external resistances that might be
effect for condition (F = 21; p = 0.001) but no significant group encountered in daily task performance. Similar to previous
main effect (F = 1.2; p = 0.31) or group by condition interaction studies, we found that muscle performance characteristics
(F = 1.2; p = 0.32) (Figure 4B). improved with both high-speed and traditional slow-speed
RT. However, novel findings were that speed-related muscle
Peak Velocity. In the first analysis (a), repeated-measures performance characteristics, such as PPV and overall VEL,
ANOVA using the change scores relative to the posttraining were most positively impacted by high-speed power training,
1RM indicated no significant group main effect (F = 2.3; p = especially at lower external resistances. In addition, there
0.11), but there was a significant condition main effect (F = were similar increases in 1RM strength obtained with both
6.1; p = 0.002) and group by condition interaction (F = 2.1; high-speed power training and traditional RT; thus, the gains
p = 0.02) (Figure 5A). To further evaluate the significant in speed-related measures with high-speed training com-
interaction, post hoc testing revealed that HSPT demon- pared to traditional RT do not come at the expense of other
strated greater VEL compared to CON at 50% 1RM muscle performance outcomes.
Since the earliest studies of muscle power in older adults, PP It has been argued that increases in the 1RM associated
(the highest power value across the spectrum of the power with RT could account for the apparent decrease in PPV after
curve) has been repeatedly shown to be a strong predictor of RT with a strengthening component (8). For example, the
functional performance (2,3,12,36) and even more important movement of heavier weights following gains in 1RM
than muscle strength (3,5,33,36). As a result, RT protocols for strength with an RT protocol would lead to the slowing of
older adults were developed with the intention of improving movement speed because of the force–velocity relationship.
peak muscle power. In 2004 and 2005, 2 studies reported that However, in this study, 1RM strength increases after 12 weeks
muscle power developed at high speeds (40% 1RM) and the of training were similar in both the high-speed and slow-
corresponding VEL at that external resistance were stronger speed training groups (;26% in both groups). This suggests
predictors of functional decline than power at 70% 1RM that speed-related increases in muscle performance in the
(a power value typical of PP values across the power curve) high-speed training group compared to the traditional RT
(5,34). These studies led us to believe that power values other were likely because of the differences in the speed of the
than the typical ‘‘peak’’ value, and factors that contribute to training protocols, not differences in 1RM improvement
power (i.e., velocity) may be important to focus on in RT between groups. Moreover, speed appears to be trainable in
studies. Although many studies have explored high-speed the older adult population. In addition, it may be misleading
power training (6,11,14,16,21,26), to our knowledge, no to simply state that PPV was not different between the high-
studies have reported the impact of this type of training on speed training group and control, given that there was no
power values across a range of external resistances. Two increase in the 1RM in the control group over 12 weeks.
studies we are aware of have examined the force and velocity Because the HSPT group significantly increased their 1RM
components of muscle power after power training (7,21); compared to control (25 vs. 0%), these 2 groups were moving
however, no studies to date have examined these compo- very different resistances when being tested at the same
percentage of the 1RM.
nents across a range of external resistances, at other than
Few studies on muscle power in older adults have
peak muscle power, and whether force at PP and PPV are
compared changes in muscle power (and its components)
differentially affected by different types of RT.
relative to the baseline 1RM (8). This was done in this study
When comparing changes in muscle performance relative
to account for potential differences in the 1RM that could
to the final posttraining 1RM, we found that both HSPT
have occurred with the different training regimens. We
and STR increased PP across a range of external resistances
initially hypothesized that training at a higher percent of the
compared to control; but, neither protocol was superior to the
1RM (80% 1RM) would improve 1RM strength to a greater
other. Thus in contrast to our initial hypothesis, PP can be
degree than at a lower percentage of the 1RM (40% 1RM);
improved similarly across a range of external resistances by
however, this did not occur in this study. As previously
performing high-velocity or slow-velocity training. Similarly,
discussed, the lack of reported change in velocity relative to
both HSPTand STR increased PPTat all power values across
the final 1RM after training (7) was largely because of
a range of external resistances compared to control, but again, training-induced increases in the 1RM; thus, speed-related
neither protocol was superior to the other. These findings in changes may not be as profound given that a heavier weight
our high-speed training group are in agreement with those of is being moved (albeit at the same percentages of 1RM).
de Vos et al. (7), who reported that the force component of In real-world situations, however, some external resistances
muscle power (at peak only) increased significantly during are fixed and would not vary with functional performance.
high-speed power training in 3 different training groups (20, For example, the force necessary to depress a gas pedal in
50, 80% 1RM). a car or rise from a chair does not change after training, so
When we compared the velocity components of muscle representing these tasks by having a person depress
power with different training regimens, however, differences a ‘‘heavier’’ brake pedal or rise from a chair wearing
began to emerge between our training groups. HSPT did not a weighted vest would not reflect the actual force necessary
increase PPVcompared to CON, again in agreement with the to accomplish the task. This, in essence, is what we measure
findings of de Vos et al. (7) but did increase PPV in HSPT when we assess changes in muscle performance after training
compared to STR. In addition, HSPT increased overall VEL relative to the final posttraining 1RM. Thus, including
compared to both STR and CON at the lowest external evaluation of performance against a fixed external resistance
resistances, ranging from 40 to 60% of the 1RM. Thus, slow- (loads based upon the baseline 1RM) might better reflect
speed training appeared to decrease speed-related compo- how speed-related changes in muscle performance are
nents of muscle performance compared to high-speed applicable to real-world situations.
training. If traditional RT results in an actual slowing of In this study, we found that changes in muscle performance
movement speed in older adults, and increased speed is across a range of external resistances produced at the same
a critical factor in function in this population (34), these 1RM as that initially obtained at baseline were improved with
findings could dramatically change the way we design and both traditional RT and high-speed power training compared
implement RT programs in older adults. to control; however, speed-related changes (PPV and VEL)
the TM
the TM
resulting from training continued to favor HSPTcompared to practical mixed-method approach including both high-speed
STR and CON. Similar to the changes in muscle performance and traditional RT.
relative to the final 1RM, the change in PP relative to the
baseline 1RM was significantly increased with HSPT and ACKNOWLEDGMENTS
STR compared to CON across the spectrum of the power Data from this study were supported by grants from the National
curve but was not significantly different from each other at Institute on Aging R03-AG025141-01 and the American
any point. Although there was no difference in the change in College of Sports Medicine Research Endowment REG 07.
PPT among all 3 groups, PPV and VEL were significantly We would like to acknowledge the contribution of Mason
increased in HSPT compared to STR and CON. Stevens, Daniel Hodgson, Jenessa Blalock, Robert Rogers,
The improvements in PPV and VEL may more accurately Shanelle Carolan, and Susan Guittar for the collection of data
reflect training-induced changes in speed characteristics of presented in this report. Results of this study do not constitute
the muscle compared to examination of changes relative to endorsement of any product by the authors or the National
the final 1RM that reflect movement against a higher absolute Strength and Conditioning Association.
mass. Again, these findings are supportive of differential
effects of RTon muscle speed characteristics. Surprisingly, we REFERENCES
also noted a strong trend in PPV (p = 0.06) and significant 1. American College of Sports Medicine. ACSM’s Guidelines for Exercise
Testing and Prescription (6th ed.). Baltimore, MD: Wolters Kluwer,
increases in VEL for our STR group compared to CON.
2000.
Delmonico et al. (8) compared changes in muscle perfor-
2. Bassey, EJ, Fiatarone, MA, O’Neill, EF, Kelly, M, Evans, WJ, and
mance after 10 weeks of traditional slow-speed RT (80–85% Lipsitz, LA. Leg extensor power and functional performance in very
1RM) in older men and women by similarly examining old men and women. Clin Sci (Lond) 82: 321–327, 1992.
changes relative to the baseline 1RM. Although they found 3. Bean, JF, Kiely, DK, Herman, S, Leveille, SG, Mizer, K, Frontera, WA,
that PPV decreased with slow-speed RT (27%), we found and Fielding, RA. The relationship between leg power and physical
performance in mobility-limited older people. J Am Geriatr Soc
a near significant increase in PPV across a range of external 50: 461–467, 2002.
resistances (range: 21 to 27%) and significant increases in 4. Borg, G. Perceived exertion as an indicator of somatic stress. Scand J
VEL. Delmonico et al. only evaluated PPV at 70% 1RM, Rehabil Med 2: 92–98, 1970.
whereas this study evaluated PPV and VEL across a range of 5. Cuoco, A, Callahan, DM, Sayers S, Frontera, WA, Bean, J, and
external resistances. However, even at 70% 1RM, the change Fielding, RA. Impact of muscle power and force on gait speed in
disabled older men and women. J Gerontol 59: 1200–1206, 2004.
in PPV with training in the STR group was still positive (17%)
6. de Vos, NJ, Singh, NA, Ross, DA, Stavrinos, TM, Orr, R, and
in this study compared to Delmonico et al. Thus, it appears Fiatarone-Singh, MA. Optimal load for increasing muscle power
from our data that even slow-velocity RT can increase speed during explosive resistance training in older adults. J Gerontol
when moving a resistance relative to the baseline 1RM. 60A: 638–647, 2005.
7. de Vos, NJ, Singh, NA, Ross, DA, Stavrinos, TM, Orr, R, and
Fiatarone-Singh, MA. The effect of explosive resistance training on
PRACTICAL APPLICATIONS force and velocity to peak power generation in older adults. J Aging
Phys Activ 16: 1–17, 2008.
High-speed power training improved speed-related muscle 8. Delmonico, MJ, Kostek, MC, Doldo, NA, Hand, BD, Bailey, JA,
performance characteristics (PPV and VEL) compared to Rabon-Stith, KM, Conway, JM, Carignan, CR, Lang J, and Hurley, BF.
traditional slow-speed RT, particularly at lower external Effects of moderate-velocity strength training on peak muscle power
and movement velocity: Do women respond differently than men?
resistances. Based on these findings, we recommend including J Appl Physiol 99: 1712–1718, 2005.
an RT protocol in older adults that emphasizes high-speed 9. Evans, WJ. Exercise strategies should be designed to increase muscle
movements at low external resistances (;40% of the 1RM). power. J Gerontol 55A: M309–M310, 2000.
To increase the velocity component of an activity that is of 10. Fiatarone, MA, Marks, E, Ryan, ND, Meredith, CN, Lipsitz, LA, and
a low percentage of the 1RM, this study clearly indicates that Evans, WJ. High-intensity strength training in nonagenarians:
Effects on skeletal muscle. JAMA 263: 3029–3034, 1990.
high-speed power training accomplishes this to an even
11. Fielding, RA, LeBrasseur, NK, Cuoco, A, Bean, J, Mizer, K, and
greater degree than traditional RT. This is important because Fiatarone-Singh, MA. High-velocity resistance training increases
people often function at percentages of the 1RM well below skeletal muscle peak power in older women. J Am Geriatr Soc
their maximum, particularly during events that require much 50: 655–662, 2002.
greater velocity than force, such as moving the lower limb 12. Foldvari, M, Clark, M, Laviolette, LC, Bernstein, MA, Kaliton, D,
Castaneda, C, Pu, CT, Hausdorff, JM, Fielding, RA, and Singh, MA.
from the gas pedal to the brake while driving or moving the
Association of muscle power with functional status in community-
limb quickly to prevent falling. The application of these dwelling elderly women. J Gerontol 55A: M192–M199, 2000.
findings to RT programs in older adults could have 13. Hagerman, FC, Walsh, SJ, Staron, SJ, Hikida, RS, Gilders, RM,
widespread public health implications by possibly increasing Murray, TF, Toma, K, and Hervey, S. Effects of high-intensity
the safety of speed-related daily activities in this population. resistance training on untrained older men. I. Strength, cardiovas-
cular, and metabolic responses. J Gerontol 55A: B336–B346, 2000.
Because most people do not use only 1 style of RT program
14. Henwood, TR, Riek, S, and Taaffe, DR. Strength versus muscle
throughout the lifetime of a fitness regimen, future studies power-specific resistance training in community-dwelling older
should explore the muscle performance responses to a more adults. J Gerontol 63: 83–91, 2008.
15. Henwood, TR and Taaffe, DR. Detraining and retraining in older 27. Newton, RU, Häkkinen, K, Häkkinen, A, McCormick, M, Volek, J,
adults following long-term muscle power or muscle strength specific and Kraemer, WJ. Mixed-methods resistance training increases
training. J Gerontol 63: 751–758, 2008. power and strength of young and older men. Med Sci Sports Exerc
16. Henwood, TR and Taaffe, DR. Improved physical performance in 34: 1367–1375, 2002.
older adults undertaking a short-term programme of high-velocity 28. Nichols, JF, Omizo, DK, Peterson, KK, and Nelson, KP. Efficacy of
resistance training. Gerontol 51: 108–115, 2005. heavy-resistance training for active women over sixty: Muscular
17. Hook, PV, Sriramoju, V, and Larsson, L. Effects of aging on actin strength, body composition, and program adherence. J Am Geriatr
sliding speed on myosin from single skeletal muscle cells of mice, rats, Soc 41: 205–210, 1993.
and humans. Am J Cell Physiol 280: C782–C788, 2001. 29. Norris, AH, Shock, NW, and Wagman, LH. Age changes in the
18. Hunter, GR, Treuth, MS, Weinsier, RL, Kekes-szabo, T, Kell, SH, maximum conduction velocity of motor fibers of human ulnar
Roth, DL, and Nicholson, C. The effects of strength conditioning on nerves. J Appl Physiol 5: 589–593, 1953.
older women’s ability to perform daily tasks. J Am Geriatr Soc 30. Orr, R, de VOS, NJ, Singh, NA, Ross, DA, Stavrinos, TM, and
43: 756–760, 1995. Fiatarone-Singh, MA. Power training improves balance in healthy
19. Lexell, J, Henricksson-Larson, K, Winblad, B, and Sjostrom, B. older adults. J Gerontol 61: 78–85, 2006.
Distribution of different fiber types in human skeletal muscles: 31. Petrella, JK, Kim, J, Tuggle, SC, Hall, SR, and Bamman, MM. Age
Effects of aging studied in whole muscle cross sections. Muscle Nerve differences in knee extension power, contractile velocity, and
6: 588–595, 1983. fatigability. J Appl Physiol 98: 211–220, 2005.
20. Lexell, J and Taylor, CC. Variability in muscle fiber areas in whole 32. Pyka, G, Lindenberger, E, Charette, S, and Marcus, R. Muscle
human quadriceps muscle: Effects of increasing age. J Anat strength and fiber adaptations to a year-long resistance training
174: 239–249, 1991. program in elderly women. J Gerontol 49: M22–M27, 1994.
21. MacAluso, A and De Vito, G. Comparison between young and older 33. Sayers, SP. High speed power training: A novel approach to
women in explosive power output and its determinants during resistance training in older men and women. A brief review and pilot
a single leg-press action after optimization of load. Eur J Appl Physiol study. J Strength Cond Res 21: 518–526, 2007.
90: 458–463, 2003. 34. Sayers, SP, Guralnik, JM, Thombs, LA, and Fielding, RA. Impact of
22. Marcell, TJ. Sarcopenia: Causes, consequences, and preventions. leg muscle contraction velocity on functional performance in older
J Gerontol 58A: 911–916, 2003. men and women. J Am Geriatr Soc 53: 467–471, 2005.
23. Marsh, AP, Miller, ME, Rejeski, WJ, Hutton, SL, and Kritchevsky, SB. 35. Sipila, S, Elorinne, E, Alen, M, Suominen, H, and Kovanen, V. Effects
Lower extremity muscle function after strength or power training in of strength and endurance training on muscled fiber characteristics
older adults. J Aging Phys Act 17: 416–443, 2009. in elderly women. Clin Phys 17: 459–474, 1997.
24. McCartney, N, Hicks, AL, Martin, J, and Webber, CE. A longitudinal 36. Suzuki, T. Bean, JF, and Fielding, RA.JF, and Fielding, RA. Muscle power
trial of weight training in the elderly: Continued improvements in of the ankle flexors predicts functional performance in community-
year 2. J Gerontol 51A: B425–B433, 1995. dwelling older women. J Am Geriatr Soc 49: 1161–1167, 2001.
25. Metter, EJ, Conwit, R, Tobin, J, and Fozard, JL. Age-associated loss 37. Taffe, DR, Duret, C, Wheeler, S, and Marcus, R. Once-weekly
of power and strength in the upper extremities in women and men. resistance exercise improves muscle strength and neuromuscular
J Gerontol 52A: B267–B276, 1997. performance in older adults. J Am Geriatr Soc 47: 1208–1214,
26. Miszko, TA, Cress, ME, Slade, JM, Covey, CJ, Agrawal, SK, and 1999.
Doerr, CE. Effect of strength and power training on physical 38. Wagman, LH and Lessee, H. Maximum conduction velocities of
function in community-dwelling older adults. J Gerontol 58: 171–175, motor fibers of ulnar nerve in human subjects of various ages and
2003. size. J Neurophysiol 15: 235–242, 1952.
the TM

Sayers 2010

Uploaded by

Copyright:

Available Formats

Sayers 2010

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sayers 2010

Uploaded by

Copyright:

Available Formats

A COMPARISON OF HIGH-SPEED POWER TRAINING

AND TRADITIONAL SLOW-SPEED RESISTANCE

VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3369

keep from falling vs. slowly getting up from a chair). Few

TABLE 1. Subject characteristics.*†

Age 74.1 6 6.4 70.1 6 7.0 72.8 6 4.1 0.38

3370 Journal of Strength and Conditioning Research

Journal of Strength and Conditioning Research

% Change: 21 % Change: 24 % Change: 27 % Change: 28 % Change: 25 % Change: 214

High-Speed Power Training in Older Adults

Journal of Strength and Conditioning Research | www.nsca-jscr.org

study. Eleven dropped out of

VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3373

training period. All outcome

Resistance Training Protocol. Vol-

3374 Journal of Strength and Conditioning Research

Journal of Strength and Conditioning Research | www.nsca-jscr.org

(group by time) repeated meas-

VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3375

Changes from Baseline

3376 Journal of Strength and Conditioning Research

Journal of Strength and Conditioning Research | www.nsca-jscr.org

VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3377

3378 Journal of Strength and Conditioning Research

Journal of Strength and Conditioning Research | www.nsca-jscr.org

VOLUME 24 | NUMBER 12 | DECEMBER 2010 | 3379

3380 Journal of Strength and Conditioning Research

You might also like