Content uploaded by Ignacio J. Chirosa
Author content
All content in this area was uploaded by Ignacio J. Chirosa on Oct 13, 2017
Content may be subject to copyright.
BRIEF REVIEW
EFFECTS OF VARIABLE RESISTANCE TRAINING ON
MAXIMAL STRENGTH:AMETA-ANALYSIS
MIGUEL A. SORIA-GILA,IGNACIO J. CHIROSA,IKER J. BAUTISTA,SALVADOR BAENA,AND
LUIS J. CHIROSA
Department of Physical Education and Sport, University of Granada, Granada, Spain
ABSTRACT
Soria-Gila,MA,Chirosa,IJ,Bautista,IJ,Baena,S,andChirosa,LJ.
Effects of variable resistance training on maximal strength: A meta-
analysis. J Strength Cond Res 29(11): 3260–3270, 2015—
Variable resistance training (VRT) methods improve the rate of
force development, coordination between antagonist and syner-
gist muscles, the recruitment of motor units, and reduce the drop
in force produced in the sticking region. However, the beneficial
effects of long-term VRT on maximal strength both in athletes and
untrained individuals have been much disputed. The purpose of
this study was to compare in a meta-analysis the effects of a long-
term ($7 weeks) VRT program using chains or elastic bands and
a similar constant resistance program in both trained adults prac-
ticing different sports and untrained individuals. Intervention effect
sizes were compared among investigations meeting our selection
and inclusion criteria using a random-effects model. The published
studies considered were those addressing VRT effects on the 1
repetition maximum. Seven studies involving 235 subjects fulfilled
the selection and inclusion criteria. Variable resistance training led
to a significantly greater mean strength gain (weighted mean dif-
ference: 5.03 kg; 95% confidence interval: 2.26–7.80 kg; Z=
3.55; p,0.001) than the gain recorded in response to conven-
tional weight training. Long-term VRT training using chains or
elastic bands attached to the barbell emerged as an effective
evidence-based method of improving maximal strength both in
athletes with different sports backgrounds and untrained subjects.
KEY WORDS one repetition maximum, elastic bands, chains,
biomechanics
INTRODUCTION
Over the past few years, strength training proto-
cols designed to optimize the efficiency and ben-
efits of training have gained popularity (20,33).
Strength training programs including variable
resistance (VR) exercises are typically performed using
accessories, such as elastic bands or chains, and machines
that allow for variation in the velocity of load displacement
and its magnitude. One of the main objectives of the use of
chains or elastic bands is to induce a high variation of stimuli
and thus provoke neural adaptations improving the different
expressions of strength, including maximal strength or the
1 repetition maximum (1RM) (3,28). These methods com-
bine the resistance generated by fixed loads (e.g., barbell and
disks) with the VR produced by elastic bands and chains
attached to the barbell. The most characteristic feature of
this training modality is that resistance directed against the
target muscle or muscle group can be varied over the range
of athletic movement (1,20). Many authors claim that this
type of resistance training reduces the mechanical disadvan-
tage of the sticking point encountered in free weight training
(2–4,33,38). The sticking point or sticking region refers to
the loss of velocity produced in external resistance exercise
and was first described by the authors of classic studies such
as Elliott et al. (18). More recently, van der Tillar and Ettema
(44) discovered that the sticking region is dependent on
loading and accounts for 35–45% of the range of movement.
The sticking region is the most inefficient stage of a joint
movement in that the muscle groups involved cannot meet
the demands of exercise when working with loads as high as
90% of the 1RM (36) or even 80% (18). In this region, move-
ment velocity decreases most likely because of compromised
neural intermuscular and intramuscular coordination, result-
ing in a reduction in the force sustained (44). The rationale
for variable resistance training (VRT) is that a greater abso-
lute external load will be supported if this neuromechanical
disadvantage is minimized by applying lower resistances
(loads lower than 85% 1RM, Table 1 indicating the loads
sustained at the end of the athletic movement) across less
efficient movement ranges (2,18). According to van den
Tillaar and Saterbakken (45), in practical terms, this means
that these movement ranges could be avoided by controlling
exercise velocity to increase the mechanical impulse of each
exercise repetition for workloads greater than 80% 1RM at
the start of the sticking region.
During a variable intrarepetition stimulus weight lifting
protocol, a load increase takes place as the barbell is moved
through the concentric phase of the range of motion, making
it increasingly more difficult to maintain a high velocity and
Address correspondence to Miguel A
´. Soria-Gila, ma88@correo.ugr.es.
29(11)/3260–3270
Journal of Strength and Conditioning Research
Ó2015 National Strength and Conditioning Association
3260
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
TABLE 1. Details of the studies included in the meta-analysis.*†
Reference n
Subject characteristics
Gender
Age (6SD)
(y)
Weight (6SD)
(kg)
Height (6SD)
(cm) Training level Sports activity
Anderson et al. (2) 44 Mixed 20 61 66.2 613.8 zTrained (4 62 y) Basketball, wrestling, and hockey
Bellar et al. (9) 11 Men 23.6 63.2 84.4 618.8 179 68.5 Untrained z
Cronin et al. (15) 40 Mixed 23.1 64.8 76.3 611 175 69 Trained (3 y) z
Ghigiarelli et al. (22) 36 Men 19.96 61.03 96.3 615 180.83 66.24 Trained American football (36 Division
1AA)
McCurdy et al. (31) 27 Men 20.63 61.33 84.79 65.84 178.89 65.46 Trained (4.8 62.7 y) Baseball (Division II)
Rhea et al. (40) 48 Men 21.4 62.1 zzTrained Athletics (NCAA) Division I
Shoepe et al. (42) 29 Mixed 19.76 61.33 66.8 611.1 168.77 610.3 Scarce training (12 mo) Free weight lifting
Reference
Training characteristics
VRT Duration Sets Repetitions Rest (s)
Intensity
PCR PVR PMR
Anderson et al. (2) Elastic bands 3 d$wk
21
37 wk 3–6 2–10 120–180 85 15 85
Bellar et al. (9) Elastic bands 2 d$wk
21
313 wk 5 5 90 85 15 85
Cronin et al. (15) Elastic bands 2 d$wk
21
310 wk 3 8–15 zzzz
Ghigiarelli et al. (22) Elastic bands and chains 4–5 d$wk
21
37 wk 5–6 4–6 zzz85
McCurdy et al. (31) Chains 2 d$wk
21
39 wk 5–7 5–10 z80–90 10–20 z
Rhea et al. (40) Elastic bands 2–3 d$wk
21
312 wk 4 10 zzz75–85
Shoepe et al. (42) Elastic bands 3 d$wk
21
324 wk 3–6 6–10 60–120 80–65 20–35 65–95
*d$wk
21
= days per week; VRT = variable resistance training; PCR = percentage constant resistance; PVR = percentage variable resistance; PMR = percentage maximum
resistance; NCAA = National Collegiate Athletic Association.
†Individual details of each study in terms of sample size (n), subject characteristics, and resistance training characteristics.
zNot defined.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 29 | NUMBER 11 | NOVEMBER 2015 | 3261
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
acceleration (11,13,19,48). When using elastic bands, suffi-
cient acceleration is needed in the early lifting stage to over-
come elastic recoil and complete the movement (19).
Contrarily to the use of bands, chains act by adding mass
(19). The magnitude of this system of masses is proportional
to the height reached by the barbell from the ground. The
gradual increase produced in external resistance causes insta-
bility, which could induce an optimal stimulus for strength
gains (32), and a high neuromuscular demand, increasing
both motor unit recruitment and rate coding (25). Such neu-
romuscular adaptations could be the consequence of
improved coordination between antagonist and synergist
muscles controlling movement (2,8,9,16,31,38). Some
authors have argued that greater muscle activation due to
stored elastic energy translates to an improved rate of force
development (RFD) (40). The resistance produced by elastic
bands or chains generates the greatest workload at the end
of the range of motion. In other words, a steady load
increase is produced through the trajectory of movement,
whereas in traditional training using free weights, this great-
est load is sustained at the onset of the concentric phase
(22,23). A further issue to consider is that elastic bands
increase resistance in a curvilinear manner, whereas chains
do so linearly because of their different physical and
mechanical properties (15,19,33,34).
The results of recent studies (2,9) assessing the efficiency
of combining elastic tension with the tension induced by free
weights in traditional back squat exercise suggest that
80–85% of the load should be provided by free weights
and 15–20% by VR. To improve peak power during explo-
sive movements when elastic bands and free weights are
used in the back squat, other authors (28,33,38,48) recom-
mend figures of 20–35% and 65–80% for VR- and free-
weight loaded resistance, respectively.
A further characteristic feature of varying resistance is that,
besides increasing velocity, it increases the eccentric stimulus
of training, and thus the strength needed to slow down or stop
the load at the end of the eccentric phase, inducing greater
myoelectric activity in the muscles (15). Researchers examin-
ing VRT using chains have also reported that this type of
training induces stimulus variations, as a consequence of the
oscillations that chains produce, which provoke better coor-
dination between agonist, synergistic, and stabilizing muscles
to control the load (11,31). Several studies (2,21,48) have
detected improvements in muscular strength and power gen-
erated in bench press and squat exercises in response to elastic
plus free weight loaded training, compared with similar train-
ing in the absence of elastic resistance. In addition, VRT
improves resistance to fatigue through the physiologic
response to an acute effect of fatigue during training (46).
Individual differences in muscle contractile properties can also
lead to different degrees of fatigue (47).
Based on the available literature, it is difficult to extract
whether the different VRT programs show true benefits in
improving muscular strength. The present meta-analysis was
designed to examine research-based information on the
effects on maximal strength, or 1RM, of a long-term VRT
program under different training conditions.
METHODS
A meta-analysis was designed following the recommenda-
tions and criteria proposed by the Cochrane Review Group
(26). Each step (article identification, filtering, eligibility
assessment, and inclusion/exclusion of a study) was per-
formed by the present authors.
Selection and Inclusion/Exclusion Criteria
All randomized controlled studies assessing the effects of
a 7-week or longer VRT intervention providing maximal
strength as the main outcome variable were identified. There
were no restrictions made on the search regarding gender,
training status, sport specialty, or body mass index.
A study was included if VRT intervention duration was
$7 weeks and involved $2 sessions per week. The former
cutoff was based on the finding that 6 weeks of resistance
training is sufficient to improve the maximal strength of the
knee extensors by 35%, as a consequence of an increase in
the motor unit firing rate (29). The number of sessions per
week was based on the findings of Rhea et al. (39). The use
of elastic bands or chains was also an inclusion criterion
although the training method (e.g., bench press or back
squat) used to improve strength was not a limitation. Only
articles providing preintervention and postintervention 1RM
data were included. Studies were excluded if designed to
treat a disorder or disease.
Articles were required to report on at least 1 subject group
undergoing VRT and to include a control group of individ-
uals training using the more traditional method (i.e., using
free weights). Also as an inclusion criterion, we considered
all valid and reliable methods commonly used to measure
maximum strength in the different studies (14,32,34).
Search Methods
The following databases were searched for articles published
before January 2014: MEDLINE, PubMed; Scopus, SPORT-
Discus, and Web of Science using the keywords bench press,
bungee weight, chain, concentric, eccentric, elastic bands,
exercise, force, free weight, load, lower limb, maximal, muscle,
neuromuscular, output, performance, power, resistance pro-
gram, rubber bands, squat, strength, traditional, training, upper
limb, variable, and velocity. Annual scientific conference
abstracts, comments, or duplicated publications of the same
study were not included. We also examined references listed
and cited in the articles identified, including review articles, to
identify additional studies. The full texts of the all the articles
selected were examined by 3 of the present authors (M.A.S.-G.,
I.J.C., and S.B.).
Statistical Analyses
Study Inclusion. To select the studies for final inclusion in the
meta-analysis, the 3 reviewers independently screened the
Variable Resistance Training
3262
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
references identified for eligibility. Abstracts were assessed
for the studies’ fulfillment of inclusion/exclusion criteria.
Study quality criteria were also considered (experimental
design, subject withdrawal, and possible conflicts of interest).
The recently developed QAREL checklist (30) was used to
evaluate the methodological quality of included interrater
reliability and agreement studies. This checklist was chosen
because it seems to be the only available quality appraisal
tool for reliability studies at the moment. Any disagreement
between reviewers was resolved by consensus.
Interstudy Heterogeneity. Variation between studies was as-
sessed in terms of the effect under investigation (i.e., maximal
strength). Effect sizes (ES) are provided as differences in
weighted means, along with the corresponding 95% confi-
dence interval (CI). To estimate interstudy heterogeneity, the
x
2
method was used with significance set at p#0.05. The
index I
2
was also determined,
where 0% indicates homogene-
ity and 100% heterogeneity (27).
Coding of Studies. Each study
was read and coded by the
main investigator according to
the following variables: descrip-
tive information including sam-
ple size, gender, age, weight,
height, training level, sports
activity,typeofVRT,extremi-
ties trained, training duration,
sets, repetitions, rest, percentage
constant resistance, percentage
VR, and percentage maximum
resistance (PMR).
Coder drift was assessed (37)
by randomly selecting 4 studies
for recoding. Per case agree-
ment was determined by divid-
ing the variables coded the
same by the total number of var-
iables. A greater mean agreement
level was observed (93%) than
the minimum accepted level
of 90%.
Effect Size. The effects of the intervention were calculated for
each study using the pretraining and posttraining mean and
SDs recorded for the main outcome measure (1RM) in the
VRTand control (conventional training) groups. The pooled
ES was estimated in terms of the change in SD produced.
When a study lacked the necessary data to estimate the SD
change, the following equation was used:
where, corr is a correlation factor that relates pretraining and
posttraining results based on the data provided by Rhea et al.
(40) (0.96 for the VRT groups and 0.97 for the control
groups).
A random-effects model was used to examine the grouped
data extracted from the different studies. The relative
strength of the intervention effect and 95% CIs for each
study were illustrated in a forest plot. The ES of the
intervention was calculated as the difference between
pretraining and posttraining 1RM mean.
Figure 1. Study selection/inclusion process.
SD change ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
½SD pre2þ½SD post2223corr2½pre;post3SD pre3SD post
r;(1)
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 29 | NUMBER 11 | NOVEMBER 2015 | 3263
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
In a separate sensitivity analysis, we determined the
contribution of each study to the overall improvement in
maximal strength detected in this meta-analysis by successively
omitting the results of each
study from the comparisons
made using the data from the
remaining studies.
All calculations were per-
formed using the RevMan
software package (Review
Manager–Version 5.2; The Co-
chrane Collaboration, 2012).
RESULTS
Study Characteristics
Seven studies providing results
for 16 subject groups met the
criteria for inclusion in our
meta-analysis (2,9,15,22,31,40,42),
(Figure 1). Publication dates were
2003–2011. An overview of the
characteristics of the 7 studies
included in this meta-analysis
is provided in Table 1. All stud-
ies selected were designed to
address the same issue although
working hypotheses differed
slightly. Some studies compared
the effects on the 1RM of train-
ing using free weights with chains (22,31), whereas others
compared several experimental groups subjected to different
VRT interventions (elastic bands or chains) with a control
group (traditional free weight
training) (15,22). In the study
by Rhea et al. (40), several
experimental groups undertak-
ing different training protocols
with chains were compared.
Another study compared the ef-
fects of training with elastic
bandsattachedtofreeweights
in bench presses and squats (42).
Subject Characteristics
The data examined were ob-
tained from 235 subjects aged
18.3–27.9 years (mean 6SD:
21.21 62.11 years) (Table 1).
Four of the 7 studies were con-
ducted only in men (10 groups)
and 3 in both men and women
(6 groups). The participants of 2
studies were untrained subjects
or had less than 12 months of
experience (4 groups). In 5 stud-
ies, subjects had experience of 2
years or longer or were trained
(12 groups). Trained subjects
Figure 2. Forest plot of the results of the meta-analysis of random effects showing the difference in mean
weighted 1RM and 95% CI detected for the bench press, leg press, back squat, and squat (5.03 kg; 95% CI:
2.26–7.80 kg; Z= 3.55; p,0.001) in trained and untrained subjects. Gray squares indicate the intervention
effect. Square sizes are proportional to the weights assigned to each study in the meta-analysis. The horizontal line
joins the lower and the upper limits of the effect at a 95% CI. The diamonds represent the subgroup mean
difference (◊) and pooled mean difference (♦). BP = bench press; LP = leg press; SQ = squat; BSQ = back
squat; EB = elastic band; CH = chain; FW = free weight; CI = confidence interval; 1 = fast-velocity group;
2 = slow-velocity group.
Figure 3. Forest plot of the results of the meta-analysis of random effects showing the difference in mean
weighted 1RM and 95% CI detected for the bench press, leg press, back squat, and squat (5.03 kg; 95%
CI: 2.26–7.80 kg; Z= 3.55; p,0.001) in upper-body training and lower-body training subjects. Gray squares
indicate the intervention effect. Square sizes are proportional to the weights assigned to each study in the meta-
analysis. The horizontal line joins the lower and the upper limits of the effect at a 95% CI. The diamonds represent
the subgroup mean difference (◊) and pooled mean difference (♦). BP = bench press; LP = leg press; SQ =
squat; BSQ = back squat; EB = elastic band; CH = chain; FW = free weight; CI = confidence interval; 1 = fast-
velocity group; 2 = slow-velocity group.
Variable Resistance Training
3264
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
TABLE 2. Results of the 7 studies included in the meta-analysis.*
Reference Exercise
1RM
Control groups Experimental groups
n
Pretreatment
(kg)
SD
(kg)
Posttreatment
(kg)
SD
(kg) ES n
Pretreatment
(kg)
SD
(kg)
Posttreatment
(kg)
SD
(kg) ES
Anderson et al. (2) BP 21 81.07 32.82 84.41 33.37 0.43 23 80.69 35.34 87.37 35.52 0.66
BSQ 21 108.19 35.61 115.28 33.70 0.86 23 105.80 33.70 121.75 35.70 1.58
Bellar et al. (9) BP 11 101.50 19.60 109.00 20.30 1.60 11 100.00 18.90 109.90 19.40 1.80
Cronin et al. (15) LP 12 122.00 34.10 118.95 33.25 0.39 14 128.00 27.50 139.14 29.90 1.30
Ghigiarelli et al. (22) BP 12 141.80 23.00 149.50 23.00 1.44 12 127.70 25.00 137.70 25.00 1.40
BP 12 129.50 15.00 138.60 14.00 2.13
McCurdy et al. (31) BP 27/2†102.65 14.42 109.09 12.98 1.85 27/2†151.85 27.12 174.26 13.47 1.52
Rhea et al. (40) BSQ 16 115.94 36.07 119.18 35.56 0.38 16 116.00 31.43 125.81 30.69 1.10
BSQ 16 122.31 39.04 131.94 36.43 1.08
Shoepe et al. (42) BP 12 56.30 30.30 66.70 27.00 1.40 12 53.60 21.00 59.30 24.50 0.77
SQ 12 66.90 16.50 88.90 23.20 2.72 12 69.30 27.00 91.40 31.90 2.27
*n = number of subjects in each group; ES = effect size; BP = bench press; BSQ = back squat; SQ = squat; LP = leg press.
†Not defined.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 29 | NUMBER 11 | NOVEMBER 2015 | 3265
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
were Division I athletes (National Collegiate Athletic Associ-
ation [NCAA]), baseball players (Division II) and American
football players (36 Division 1AA players).
Variable Resistance Training
Mean training program duration was 12 65 weeks (range,
7–24 weeks). From 2 to 5 training sessions were conducted
per week, with a mean of 3 61 per week. Training took the
form of upper limb exercise (bench press) in 4 studies (10
groups), lower limb exercise (back squat) in 1 study (2
groups), and both upper and lower limb training (bench
press and back squat) in 2 studies (4 groups). Chains
attached to the barbell in the bench press were used in 2
studies (2 groups), and elastic bands attached to the barbell
in bench press or back squat exercise were used in 5 studies
(6 groups each).
Publication Bias and Interstudy Heterogeneity
A scatter plot of intervention effect (1RM) against the study
size showed a funnel-shaped symmetric distribution indicat-
ing no publication bias. The treatment effect, or 1RM,
yielded the values x
2
(10) = 27.21; p= 0.002; I
2
=63%
indicating moderate interstudy heterogeneity.
Maximal Strength (One Repetition Maximum)
The mean strength gain produced was greater in the subjects
undertaking long-term VRT, the ES being 1.42 60.51 ex-
pressed as the mean 6SD (difference in the weighted mean
1RM was 5.03 kg; 95% CI: 2.26–7.80 kg; Z= 3.55; p,0.001;
Figure 2) than in those subjected to a conventional resistance
training program, with an ES of 1.24 60.71. Furthermore,
a subgroup analysis by training status indicated a significantly
better 1RM gain in response to VRT, ES = 1.35 60.43, vs.
traditional training, ES = 0.98 60.56, for trained subjects
(pooled estimate = 6.12 kg; 95% CI: 2.43, 9.80 kg; Z= 3.25;
p= 0.001; Figure 2). However, in untrained subjects, the
greater improvement produced in the 1RM with VRT, ES
=1.6260.77, compared with conventional training, ES =
1.91 60.71 was nonsignificant (pooled estimate = 2.56 kg;
95% CI: 20.55, 5.68 kg; Z= 1.61; p= 0.11; Figure 2). Another
subgroup analysis revealed that for upper extremity training,
significant differences in 1RM gains existed between the VRT
program, ES = 1.38 60.57, and traditional training program,
ES = 1.36 60.48 (pooled estimate = 3.99 kg; 95% CI: 0.92,
7.06 kg; Z= 2.54; p= 0.01; Figure 3). Similarly, for lower limb
training, VRTalso led to a significantly better improvement in
the 1RM, ES = 1.47 60.49, than conventional training, ES =
1.09 60.96 (pooled estimate = 6.07 kg; 95% CI: 0.95, 11.20
kg; Z= 2.32; p= 0.02; Figure 3). In Table 2, we provide details
of the effects of the VRT program detected in each study.
Sensitivity
In each comparison (preintervention vs. postintervention) in
which the results of 1 study were omitted, no significant
differences were detected (p,0.001) in each case indicating
the significant contribution of all the studies to the overall
strength gains observed.
DISCUSSION
In this meta-analysis, we compared the effects of traditional
vs. VRT on the adaptive response produced in terms of
maximal strength. The studies meeting the selection and
inclusion criteria for the meta-analysis were those by Cronin
et al. (15), Anderson et al. (2), Ghigiarelli et al. (22),
McCurdy et al. (31), Rhea et al. (40), Bellar et al. (9), and
Shoepe et al. (42). Participants were either untrained (with
under 12 months’ experience in strength training) or trained
(longer than 2 years’ experience). Our results indicate that
VRT over at least 7 weeks ($2 sessions per week) leads to
a significantly greater strength gain (p,0.001) than that
produced in response to a traditional strength training pro-
gram. When subjects were stratified according to training
status, trained individuals achieved a significantly greater
strength gain with the VRT than the traditional training pro-
gram (p= 0.001). However, the strength gains observed for
the nontrained subjects undertaking a VRT program vs. a tra-
ditional program did not vary significantly (p= 0.11). When
stratified according to the extremities trained, for both the
lower and upper limbs, VRT gave rise to significantly better
gains in 1RM than conventional training (p#0.02).
According to the Rhea scale (38) used to determine the
magnitude of the ES in a study comparing the effects of resis-
tance training as a function of training status, in trained sub-
jects who undertook a VRT (ES = 1.35 60.43) vs.
conventional training program (ES = 0.98 60.56), the ES
was moderate, although sufficient for a significant difference
to exist between the 2 groups. This indicates that in individuals
with more experience in resistance exercises such as the bench
press and back squat, $7 weeks of VRT is an effective stimulus
for them to show a performance peak during training.
However, in our study, a moderate ES was also observed in
untrained subjects undertaking both a VRT program (ES =
1.62 60.77) or conventional training program (ES = 1.91 6
0.71). It should be noted that subjects labeled in our study as
“trained subjects,” would according to Rhea classification (38),
be considered “recreationally trained” given they had more
than 1 year of experience but a training duration of less than
5 years.
It also remains unclear whether a VRT program of
duration under 7 weeks or longer than 12 weeks would be
adequate for athletes to develop sufficient neural and
muscular adaptations in a short time span to improve their
1RM while also continuing to improve their 1RM over the
ensuing weeks. Only one of the studies reviewed here (42)
examined a VRTprogram lasting longer than 12 weeks. This
24-week intervention in untrained subjects produced no sig-
nificant impact on 1RM.
Evidence has only recently started to mount indicating that
VRT leads to a greater RFD and muscular power than the
more conventional form of resistance training (42). The find-
ings of the latter study suggest that during VRT, the training
impulse and muscle activation achieved on completion of
Variable Resistance Training
3266
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
each repetition are enhanced. According to Shoepe et al. (42),
this means that the lifter develops greater force in the final
portion of the concentric phase. Despite this, no significant
strength differences were detected in response to a 24-week
program of traditional training and VRT training combining
elastic and free weight loading in subjects with limited RT
experience. McCurdy et al. (31) also noted improved strength
gains in individuals undertaking the VRT program but again
differences with respect to controls were not significant.
These authors attributed the strength gains observed to the
different stability involved in the variable and conventional
training protocols, such that the neuromuscular activity
required for a strength improvement depends on the stabili-
zation needed to control resistance (5). Accordingly, for
a more unstable load, greater neuromuscular activation is nec-
essary and force production is significantly reduced (6). Our
findings confirm those of the study by Shoepe et al. (42) in
that subjects unaccustomed to free weight training showed no
significant differences when comparing the effects of conven-
tional and VRT. Thus, an optimal level of stability could be
a prerequisite for an improvement in maximal strength.
Chain-loaded VRT is slightly more unstable than free weight
resistance training (31). Consequently, once an individual
becomes accustomed to VRT and acquires more neuromus-
cular control, VRT can be an optimal stimulus to develop the
different expressions of strength.
The general trend detected in this meta-analysis is in line
with the findings of the study by Anderson et al. (2), in which
significant 1RM improvements were obtained both in the
bench press and squat. Participants of this study were trained
athletes who showed no muscle cross-section increase at the
end of the training period, suggesting improvements at the
neural level. Variable resistance training emerged as a beneficial
strategy for trained athletes, offering new stimuli inducing
fitness adaptations. The strength gains produced in these ath-
letes could also be attributed to increased muscle tension in
the more mechanically productive regions of the range of
movement, accompanied by reduced loading in the less effi-
cient sticking region. According to Anderson et al. (2), during
traditional free weight training, the barbell gains velocity dur-
ing muscle shortening until the sticking region. In the latter
study, subjects executing VRT achieved approximately 10%
less resistance in the lower region of the range of movement
and 10% more resistance toward the top, or end, of the ath-
letic movement. Acceleration remained constant over a long
period within each repetition, determining that deceleration is
reduced in VRT. Bellar et al. (9) argue that another method of
modifying resistance during a traditional resistance exercise is
to add elastic resistance. Thus, variable-resistance loading dur-
ing the bench press makes the lifting movement no longer
isoinertial. The percentage-load variation produced by elastic
bands here was 15%, and this modified the strength produc-
tion pattern during lifting. This type of variable stimulation
could be responsible for beneficial neural adaptations. In the
subject populations entered in our meta-analysis, resistance
exercise led to improved performance in terms of maximal
strength gains (2,9,15,22,31,40,42).
Elastic recoil during eccentric contraction in VRT training
may differently challenge the neuromuscular system during
each repetition (2). Ha
¨kkinen et al. (24) reported increased
electromyographic activity and a controlled increase in veloc-
ity during eccentric actions. In another study, Cronin et al.
(15) concluded that VRT using elastic bands attached to
a jump squat machine induced greater electromyographic
activity in eccentric contractions compared with traditional
training methods. Anderson et al. (2) proposed that greater
muscle fiber recruitment and stimulation during the eccentric
portion of each repetition may bring about greater neuromus-
cular adaptations and/or type IIx fiber recruitment with VRT
than with free weights alone. This explanation offered by
Anderson et al. (2) is consistent with the preferential recruit-
ment of high-threshold motor units during high-force eccen-
tric contractions reported by Nardone et al. (35).
In the study by Ghigiarelli et al. (22), significant maximal
strength increases were observed in VR compared with tradi-
tional resistance-trained individuals, regardless of the use of
chains or elastic bands. Wallace et al. (48) observed that add-
ing elastic-loaded resistance to free weight training in the back
squat led to maximal strength and power improvements when
working with loads approaching 85% of the 1RM.
According to Cronin et al. (15), the ability to quickly com-
plete a stride and return to the starting position or move in
another direction is a determining factor for success in sports,
such as squash, badminton, tennis, and fencing, among others.
In their study, subjects undertaking VRT using elastic bands
on the leg press machine showed improvement in the move
toward the stride, especially in the last part of the eccentric
phase. These subjects were able to complete a stride faster
than their peers who trained on the same machine in the
traditional way. Thus, it seems that VRT serves to improve
the transition from eccentric to concentric phase exercise, and
thus, shortens the stretch-shortening cycle, which would
potentiate the concentric phase (12) and expedite the stride.
Despite an increased prevalence of VRT programs using
heavy chains and elastic bands, some studies have generated
contradictory results (4,10,48), whereas others have found
that VRT programs offer promising results in the long
term (15,22).
Variable resistance training using heavy chains modifies
the kinetics of the barbell for all movement ranges,
increasing the mechanical advantage of the athlete’s move-
ment (4,10,38). However, because of the gradual resistance
reduction at the end of the eccentric phase, the time taken to
reach maximum acceleration (at the start of the concentric
phase) decreases in that range of movement zone causing
neural adaptation (22,31). McCurdy et al. (31) identified the
individual range of movement as an important factor to con-
sider when quantifying the workload. Behm and Sale (7)
described the user’s intention to displace the barbell as rap-
idly as possible as the main force driving neural adaptations
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 29 | NUMBER 11 | NOVEMBER 2015 | 3267
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
of muscular power and strength. Neuromuscular adaptations
are specific to the nature of the training load (43). Thus, it
has been proposed that the different characteristics of load
distribution during VRT affect muscle recruitment patterns
(2). In the study by Anderson et al. (2), subjects in the VRT
group were able to complete the prescribed exercise sets,
whereas some of the control group subjects had to pause
for 5–10 seconds between some repetitions to complete
a set.
In their study, Rhea et al. (40) noted maximal strength gains
when they compared subjects in whom VRT involved high-
velocity movements (0.6–0.8 m$s
21
) with subjects in the tra-
ditional training group working at slower velocities (0.2–0.4
m$s
21
). These results support the idea that the RFD can be
improved through VRT training using elastic bands
(2,15,16,22). Wallace et al. (48) suggested that the RFD
increase could correspond to an earlier phase in which the
peak velocity is reached in VRT. This is because as resistance
progressively increases with the mechanical advantage, higher
levels of force are generated during the concentric phase just
at the moment when muscles approach their optimal length-
tension relationship (17). A further factor inducing an increase
in RFD is a shorter muscle tissue stretch-shortening cycle
(40). The muscle is able to store elastic energy during the
eccentric phase of movement and then releases this energy
as kinetic energy during the concentric phase of the lift (15).
According to Rhea et al. (40), when the time taken needed to
reach maximal force is not limited, strength relates more to
activation of muscle mass with some relationship to synchro-
nization. Athletes using elastic bands as the VRT stimulus
showed both improved muscular strength and power, most
likely because of simultaneously improved motor unit syn-
chronization and coding velocity, although this needs to be
confirmed in further work (40).
The finding of this meta-analysis that VRT training using
chains or elastic bands leads to strength gains has obvious
implications to be considered by coaches and specialists in
sport sciences. This new training modality enables both elite
athletes and untrained individuals to more rapidly and
efficiently achieve adaptations in their functional capacity
than the more traditional resistance training methods.
As a limitation to this meta-analysis, we should mention
that many studies were excluded because of the strict
inclusion criteria established. Similarly, because of missing
data in some of the selected studies, a correlation factor
(Equation 1) had to be calculated from the data provided by
Rhea et al. (40). A further limitation was the possible effects
of publication bias (41). Despite these limitations, this meta-
analysis provides an overview of the research in this field and
offers an explanation based on the scientific literature of the
benefits of the use of VRT to increase maximal strength.
PRACTICAL APPLICATIONS
This meta-analysis provides research-based data supporting
the benefits of VRT using chains or elastic bands as an
effective strategy to increase maximal strength (1RM) in
athletes of different sports disciplines. Thus, VRT could be
used as a complement to traditional training to vary the
athletic stimulus once the user has adapted to the previous
stimulus, leading to faster training-induced adaptations.
This training modality could help avoid overload during
the range of athletic movement and may therefore be used
throughout a sport’s season to gradually improve a competi-
tion skill. It is also a useful tool to strengthen certain muscle
groups while subjecting injured muscles to lower resistances
during a rehabilitation process.
Variable resistance training is an economic simple strategy
for use with barbells in exercises such as the bench press or
back squat. The chains or elastic bands are quick to attach
and unattach meaning that strength conditioning coaches
can readily prescribe a different exercise after a variable-
resistance exercise without wasting valuable training time.
Our results indicate that training status affects the impacts of
conventional and VRT. For untrained subjects, we would not
recommend VRT, because similar strength gains are pro-
duced with traditional free weight training. In contrast, in
trained individuals, VRTwill lead to improved strength gains
over traditional training. This type of protocol would be
ideal in adults with training experience to achieve stimulus
variations and thus avoid plateaus in their physical fitness.
This issue should be borne in mind by strength conditioning
experts and coaches to better design training regimens.
Based on our findings, it would also seem that both upper
and lower limb VRT produces greater adaptations than
conventional free weight training, indicating similar effects of
this training form on both halves of the body.
Our findings provide direction for future studies designed
to determine whether other percentages of VR work and/or
PMR will produce the same 1RM adaptations or whether
single-joint exercises will give rise to similar results as
multijoint actions. Future research efforts should also explore
whether the impacts of VRT are reduced with training
duration and establish the minimum period for VRT to
produce the strength gains detected here.
ACKNOWLEDGMENTS
The authors thank Pedro Femia Marzo for help with the
data analysis and useful comments. This study received no
funds from an external source. The results of this study do
not represent the endorsement of any product by the
authors or National Strength and Conditioning Association.
This article was based on data from a PhD thesis in
Biomedicine (Universidad de Granada, Spain) by M.A.
Soria-Gila.
REFERENCES
1. Aboodarda, SJ, Shariff, MAH, Muhamed, AMC, Ibrahim, F, and
Yusof, A. Electromyographic activity and applied load during high
intensity elastic resistance and nautilus machine exercises. J Hum
Kinet 30: 5–12, 2011.
Variable Resistance Training
3268
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
2. Anderson, CE, Sforzo, GA, and Sigg, JA. The effects of combining
elastic and free weight resistance on strength and power in athletes.
J Strength Cond Res 22: 567–574, 2008.
3. Baker, DG and Newton, RU. Effect of kinetically altering
a repetition via the use of chain resistance on velocity during the
bench press. J Strength Cond Res 23: 1941–1946, 2009.
4. Baker, D and Newton, RU. Methods to increase the effectiveness of
maximal power training for the upper body. Strength Cond J 27: 24–
32, 2005.
5. Behm, DG and Anderson, KG. The role of instability with resistance
training. J Strength Cond Res 20: 716–722, 2006.
6. Behm, DG, Leonard, AM, Young, WB, Bonsey, WAC, and
MacKinnon, SN. Trunk muscle electromyographic activity with
unstable and unilateral exercises. J Strength Cond Res 19: 193–201,
2005.
7. Behm, DG and Sale, DG. Intended rather than actual movement
velocity determines velocity-specific training response. J Appl Physiol
(1985) 74: 359–368, 1993.
8. Behm, DG and Sale, DG. Velocity specificity of resistance training.
Sports Med 15: 374–388, 1993.
9. Bellar, DM, Muller, MD, Barkley, JE, Kim, C, Ida, K, Ryan, EJ,
Bliss, MV, and Glickman, EL. The effects of combined elastic- and
free-weight tension vs. free-weight tension on one-repetition
maximum strength in the bench press. J Strength Cond Res 25: 459–
463, 2011.
10. Berning, JM, Coker, CA, and Adams, KJ. Using chains for strength
and conditioning. Strength Cond J 26: 80–84, 2004.
11. Berning, JM, Coker, CA, and Briggs, D. The biomechanical and
perceptual influence of chain resistance on the performance of the
olympic clean. J Strength Cond Res 22: 390–395, 2008.
12. Bosco, C, Komi, PV, and Ito, A. Pre-stretch potentiation of human
skeletal-muscle during ballistic movement. Acta Physiol Scand 111:
135–140, 1981.
13. Coker, CA, Berning, JM, and Briggs, DL. A preliminary
investigation of the biomechanical and perceptual influence of chain
resistance on the performance of the snatch. J Strength Cond Res 20:
887–891, 2006.
14. Cormie, P, McGuigan, MR, and Newton, RU. Developing maximal
neuromuscular power part 2-training considerations for improving
maximal power production. Sports Med 41: 125–146, 2011.
15. Cronin, J, McNair, PJ, and Marshall, RN. The effects of bungy
weight training on muscle function and functional performance.
J Sports Sci 21: 59–71, 2003.
16. Ebben, WP and Jensen, RL. Electromyographic and kinetic analysis
of traditional, chain, and elastic band squats. J Strength Cond Res 16:
547–550, 2002.
17. Edman, KAP and Komi, PV. Contractile Performance of Skeletal Muscle
Fibres, Blackwell Science Ltd, Oxford, UK: PV Komi, 2008.
18. Elliott, BC, Wilson, GJ, and Kerr, GK. A biomechanical analysis of
the sticking region in the bench press. Med Sci Sports Exerc 21: 450–
462, 1989.
19. Frost, DM, Cronin, J, and Newton, RU. A biomechanical evaluation
of resistance fundamental concepts for training and sports
performance. Sports Med 40: 303–326, 2010.
20. Garcia-Lopez, D, Hernandez-Sanchez, S, Martin, E, Marin, PJ,
Zarzosa, F, and Herrero, AJ. Free-weight augmentation with elastic
bands improves bench-press kinematics in professional rugby
players. J Strength Cond Res 2014. Epub ahead of print.
21. Garcia-Lopez, D, Herrero, AJ, Gonzalez-Calvo, G, Rhea, MR, and
Marin, PJ. Influence of “in series” elastic resistance on muscular
performance during a biceps-curl set on the cable machine.
J Strength Cond Res 24: 2449–2455, 2010.
22. Ghigiarelli, JJ, Nagle, EF, Gross, FL, Robertson, RJ, Irrgang, JJ, and
Myslinski, T. The effects of a 7-week heavy elastic band and weight
chain program on upper-body strength and upper-body power in
a sample of division 1-AA football players. J Strength Cond Res 23:
756–764, 2009.
23. Gonzalez-Badillo, JJ and Sanchez-Medina, L. Movement velocity as
a measure of loading intensity in resistance training. Int J Sports Med
31: 347–352, 2010.
24. Ha
¨kkinen, K, Alen, M, and Komi, PV. Changes in isometric force-
time and relaxation-time, electromyographic and muscle-fiber
characteristics of human skeletal-muscle during strength training
and detraining. Acta Physiol Scand 125: 573–585, 1985.
25. Heckman, CJ and Enoka, RM. Motor unit. Compr Physiol 2: 2629–
2682, 2012.
26. Higgins, JPT and Green, S. Cochrane Handbook for Systematic Reviews
of Interventions Version 5.1.0 (updated March, 2011). The Cochrane
Collaboration, 2011. Available at: www.cochrane-handbook.org/.
Accessed January 18, 2012.
27. Higgins, JPT, Thompson, SG, Deeks, JJ, and Altman, DG. Measuring
inconsistency in meta-analyses. Br Med J 327: 557–560, 2003.
28. Joy, JM, Lowery, RP, Oliveira de Souza, E, and Wilson, JM. Elastic
bands as a component of periodized resistance training. J Strength
Cond Res 2013. Epub ahead of print.
29. Kamen, G and Knight, CA. Training-related adaptations in motor
unit discharge rate in young and older adults. J Gerontol A Biol Sci
Med Sci 59: 1334–1338, 2004.
30. Lucas, NP, Macaskill, P, Irwig, L, and Bogduk, N. The development
of a quality appraisal tool for studies of diagnostic reliability
(QAREL). J Clin Epidemiol 63: 854–861, 2010.
31. McCurdy, K, Langford, G, Ernest, J, Jenkerson, D, and Doscher, M.
Comparison of chain- and plate-loaded bench press training on
strength, joint pain, and muscle soreness in division II baseball
players. J Strength Cond Res 23: 187–195, 2009.
32. McCurdy, K, Langford, G, Jenkerson, D, and Doscher, M. The
validity and reliability of the 1RM bench press using chain-loaded
resistance. J Strength Cond Res 22: 678–683, 2008.
33. McMaster, DT, Cronin, J, and McGuigan, M. Forms of variable
resistance training. Strength Cond J 31: 50–64, 2009.
34. McMaster, DT, Cronin, J, and McGuigan, MR. Quantification of
rubber and chain-based resistance modes. J Strength Cond Res 24:
2056–2064, 2010.
35. Nardone, A, Romano, C, and Schieppati, M. Selective recruitment of
high-threshold human motor units during voluntary isotonic
lengthening of active muscles. J Physiol 409: 451–471, 1989.
36. Newton, RU, Murphy, AJ, Humphries, BJ, Wilson, GJ, Kraemer, WJ,
and Hakkinen, K. Influence of load and stretch shortening cycle on
the kinematics, kinetics and muscle activation that occurs during
explosive upper-body movements. Eur J Appl Physiol Occup Physiol
75: 333–342, 1997.
37. Orwin, R. The Handbook of Research Synthesis. H. Cooper and L.V.
Hedges, eds. New York, NY: Russell Sage Foundation, 1994. pp.
139–162.
38. Rhea, MR. Determining the magnitude of treatment effects in
strength training research through the use of the effect size.
J Strength Cond Res 18: 918–920, 2004.
39. Rhea, MR, Alvar, BA, Burkett, LN, and Ball, SD. A meta-analysis to
determine the dose response for strength development. Med Sci
Sports Exerc 35: 456–464, 2003.
40. Rhea, MR, Kenn, JG, and Dermody, BM. Alterations in speed of
squat movement and the use of accommodated resistance among
college athletes training for power. J Strength Cond Res 23: 2645–
2650, 2009.
41. Rosenthal, R. Meta-Analytic Procedures for Social Research. Thousand
Oaks, CA: Sage Publications, Inc, 1991.
42. Shoepe, TC, Ramirez, DA, Rovetti, RJ, Kohler, DR, and
Almstedt, HC. The effects of 24 weeks of resistance training with
simultaneous elastic and free weight loading on muscular
performance of novice lifters. J Hum Kinet 29: 93–106, 2011.
Journal of Strength and Conditioning Research
the
TM
|
www.nsca.com
VOLUME 29 | NUMBER 11 | NOVEMBER 2015 | 3269
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
43. Staron, RS, Leonardi, MJ, Karapondo, DL, Malicky, ES,
Falkel, JE, Hagerman, FC, and Hikida, RS. Strength and skeletal-
muscle adaptations in heavy-resistance-trained women after
detraining and retraining. J Appl Physiol (1985) 70: 631–640,
1991.
44. van den Tillaar, R and Ettema, G. The “sticking period” in
a maximum bench press. J Sports Sci 28: 529–535, 2010.
45. van den Tillaar, R and Saterbakken, A. The sticking region in three
chest-press exercises with increasing degrees of freedom. J Strength
Cond Res 26: 2962–2969, 2012.
46. Walker, S, Hulmi, J, Wernbom, M, Nyman, K, Kraemer, W,
Ahtiainen, J, and Ha
¨kkinen, K. Variable resistance training promotes
greater fatigue resistance but not hypertrophy versus constant
resistance training. Eur J Appl Physiol 113: 2233, 2013.
47. Walker, S, Peltonen, H, Avela, J, and Hakkinen, K. Neuromuscular
fatigue in young and older men using constant or variable resistance.
Eur J Appl Physiol 113: 1069–1079, 2013.
48. Wallace, BJ, Winchester, JB, and McGuigan, MR. Effects of elastic
bands on force and power characteristics during the back squat
exercise. J Strength Cond Res 20: 268–272, 2006.
Variable Resistance Training
3270
Journal of Strength and Conditioning Research
the
TM
Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
