Sports Med (2014) 44:1393–1402
DOI 10.1007/s40279-014-0220-8
SYSTEMATIC REVIEW
Efficiency of Jumping Exercise in Improving Bone Mineral
Density Among Premenopausal Women: A Meta-Analysis
Renqing Zhao • Meihua Zhao • Liuji Zhang
Published online: 1 July 2014
Ó Springer International Publishing Switzerland 2014
Abstract
Background Jumping exercise is frequently regarded as
an optimum strategy for increasing pubertal bone growth,
but its role in promoting or preserving adult bone mineral
density (BMD) is still undefined.
Objectives This meta-analysis aimed to evaluate the
evidence for the influence of jumping exercise on BMD in
premenopausal women and to define the effectiveness of
high-impact exercise in improving or maintaining female
bone health.
Methods We searched MEDLINE, PubMed, EMBASE,
SPORTDiscus, Google Scholar and BIOSIS up to 1 Sep-
tember 2013 for jumping exercise influence on BMD in
premenopausal women. The search terms used were
‘jumping’, ‘skipping’, ‘brief exercise’, ‘high impact’, ‘bone
density’, ‘BMD’, ‘femoral neck’, ‘lumbar spine’, and
‘trochanter’, and the search was limited to females. Six
papers met the search criteria.
Results Six studies on BMD in the femoral neck
(Q = 2.63, p = 0.854, I2 = 0.0 %), trochanter (Q = 2.10,
p = 0.10, I2 = 0.0 %) and lumbar spine (Q = 1.17,
p = 0.979, I2 = 0.0 %) were highly homogenous in
determining skeletal responses to jumping exercise.
Jumping exercise significantly increased BMD in the
femoral neck {weighted mean difference (WMD) [fixed
effect] = 0.017 g/cm2, 95 % confidence interval (CI)
Electronic supplementary material The online version of this
article (doi:10.1007/s40279-014-0220-8) contains supplementary
material, which is available to authorized users.
R. Zhao (&)
M. Zhao
L. Zhang
College of Physical Education and Health Sciences,
Zhejiang Normal University, 688 Yingbin Road, Jinhua,
Zhejiang 321004, China
e-mail: renzhao@zjnu.cn
0.014–0.20, p \ 0.001} and trochanter (WMD [fixed
effect] = 0.021, 95 % CI 0.018–0.024, p \ 0.001). How-
ever, the lumbar spine seemed to benefit less from such
high-impact exercise (p = 0.181). Visual inspection of the
plots implicated some degree of asymmetry, indicating a
slightly positive treatment effect at the femoral neck and
trochanter sites.
Conclusions Based on meta-analysis of existing studies,
the sensitivity of skeletal response to jumping exercise in
premenopausal women is significant and site-specific, with
significant benefit from high-impact exercise noted, espe-
cially at the hip.
1 Introduction
Osteoporosis is a classic aged-related skeletal disorder
characterized by compromised bone strength predisposing
a person to high risk of fracture [1]. Osteoporosis is a major
public health problem both in developed and developing
countries. It is estimated that approximately 30 million
American women suffered from osteoporosis in 2002, and
this population is expected to increase substantially in the
future [2]; in China, there are approximately 70 million
elderly osteoporotic women, along with 30 million women
experiencing sustained bone loss [3]. Osteoporosis-related
medical care causes a heavy burden both on society and on
families.
The hips and spine are the most common sites at high
risk for facture. The absolute number of osteoporosis-
associated fractures is estimated to be approximately
700,000 in vertebrae and 300,000 in hips [4]. Given that
approximately 0.25–1 % of bone mineral density (BMD) is
lost per year in premenopausal women [5], if proper pre-
vention strategies are not addressed, considerable health
123
1394
care and social costs will be incurred. Identification of
effective strategies for osteoporosis prevention is critical,
especially for premenopausal women. Based on our current
understanding of the pathology of osteoporosis, it is rec-
ommended that preventing osteoporosis may be addressed
by reducing age-related bone loss in the premenopausal
years [6].
Compelling evidence has indicated that regular exercise,
especially aerobic and weight-bearing exercise, positively
affects bone metabolism and significantly improves bone
health in premenopausal women [7, 8]. The favorable
effect of regular exercise on the skeleton stems from the
physical strain and weight-bearing that promotes bone
modeling and remodeling and consequently maintains bone
mineralization [9, 10]. Therefore, regular exercise is widely
accepted as an optimal way to stimulate bone formation
and reduce the rate of bone loss in premenopausal women.
Due to time constraints, however, regular physical
activity attracts less attention among premenopausal
women who are at risk of developing osteoporosis, even
though it is effective in promoting and maintaining BMD.
Brief exercise, such as jumping, is widely accepted as an
exercise intervention for improving bone health and
reducing bone loss. Several lines of evidence have impli-
cated significant and positive changes in BMD after a
jumping exercise intervention [11–16], especially at the
femoral neck [11–19]. Some studies have reported jumping
exercise promoting lumbar spine BMD [11–15, 17, 18],
while others did not find such favorable effects and some
even reported negative outcomes [16, 19]. Wide variation
existed in the sample sizes, training frequency and intensity
in those exercise programs; therefore, it is necessary to
analyze the positive and negative outcomes and employ a
meta-analysis to reach some general conclusions regarding
the existing body of research. Based on previous research,
this paper aimed to systematically review the studies
sharing the same outcome of interest and analyze the
overall findings regarding the efficacy of jumping exercise
for increasing or maintaining BMD at the hip and spine in
premenopausal women.
2 Methods
2.1 Search Strategy
Following PRISMA recommendations [20], a systematic
review of published papers related to the influence of
jumping exercise on BMD in premenopausal women was
conducted. The inclusion criteria are provided in Table 1;
in general, the studies were controlled trials or randomized
controlled trials (RCTs). The population of interest inclu-
ded healthy premenopausal women who were not
123
R. Zhao et al.
performing regular exercise (no more than 2.5 h per week)
prior to study enrollment. We included studies with simple
jumping exercise interventions lasting at least half a year,
considering that bone remodeling, repairing, and ion met-
abolic activity require approximately that length of time
[21, 22].
We searched MEDLINE, PubMed, EMBASE, SPORT-
Discus, Google Scholar and BIOSIS databases through 1
September 2013 for the influence of jumping exercise on
BMD in premenopausal women. The search terms used
were ‘jumping’, ‘skipping’, ‘brief exercise’, ‘high impact’,
‘bone density’, ‘BMD’, ‘femoral neck’, ‘lumbar spine’, and
‘trochanter’, and the search was limited to females (see
electronic supplementary material [ESM], Appendix 1).
Manual searches of key journals and reference lists were
included, along with structured and comprehensive data-
base searching. Six papers met the predetermined search
criteria [11–16]. Two independent reviewers obtained and
assessed the resulting full-text-version papers.
Brief exercise is defined as short-duration exercise, and
is brief and accessible for most of the population. Brief but
regular impact exercise, such as jumping, can significantly
increase hip and spine BMD [11, 13]. Given that the
effectiveness of regular exercise is influenced by compli-
ance, it is of great public health importance to determine
the effectiveness of a brief exercise intervention.
2.2 Quality Assessment
The methodological quality of studies was assessed using
the method described by Jadad et al. [23]. This widely-
utilized instrument is a three-item questionnaire that pro-
vides an assessment of bias, specifically randomization
(0–2 points), blinding (0–2 points), and withdrawal/dropout
(0–1 point). All questions are designed to elicit yes or no
answers; the total number of points possible ranges from 0
to 5.
2.3 Outcomes
The primary outcomes for this study were absolute changes
in BMD and its standard deviation (SD) at the femoral
neck, trochanter, and lumbar spine, as assessed by dual-
energy X-ray absorptiometry (DXA). Some studies did not
report post-intervention values, which are calculated based
on the baseline and follow-up measurements in the exercise
and control groups. For those reports that did not present
original change score SDs, these were calculated using a
95 % confidence interval (CI) if available. The final data
did not include those studies in which only relative change
values or graphical presentations of change scores were
available. Other outcomes calculated included the weigh-
ted mean difference (WMD) between the exercise and
Jumping Exercise Impact on Bone Health in Postmenopausal Women
Table 1 Criteria for inclusion of studies in the meta-analysis
Inclusion Description
criterion
Study design Controlled and randomized controlled trials
Population of Healthy premenopausal women (age range
interest 18–50 years) without a history of medicine
administration or disease experience affecting
bone metabolism
Exercise Subjects without regular exercise (no more than
experience 2.5 h per week) prior to study enrollment
Intervention High-impact and short-duration jumping exercise
intervention at least lasting for 6 months
Comparison Exercise intervention compared with non-exercise
control or sham exercises
Outcome Data available of changes in BMD at femoral
measurement neck, trochanter and lumbar spine determined by
DXA at pro- and post-intervention respectively
BMD bone mineral density, DXA dual-energy X-ray absorptiometry
control groups and the pooled SDs of the mean difference
between the exercise and control groups. The equations
used to calculate these values are provided in the ESM,
Appendix 2.
2.4 Data Extraction
All data were extracted by two reviewers (MZ and LZ)
who independently reviewed every selection for accuracy
and consistency. Duplicate published literature was inclu-
ded only once to ensure that no duplicate data were
reviewed in this study. The final selected studies were
reviewed by an expert on exercise and BMD to ensure
thoroughness and completeness.
2.5 Risk of Bias
Because selective outcomes were used in data reporting,
the included studies were considered to be at unclear risk
for bias. Therefore, funnel plot inspection was applied to
examine the risk of publication bias. Funnel plots are
simple scatter plots of the treatment effects estimated from
individual studies against a measure of the study size. In
the absence of bias, the plot should resemble an inverted
symmetrical funnel. Publication bias may lead to asym-
metrical funnel plots. STATA version 12 (StataCorp,
College Station, TX, USA) was used to perform the meta-
analysis and production of graphics.
2.6 Data Analysis
We used WMD methods to combine study effect size
estimates. This method pools the effect estimate
1395
representing a weighted average of all included study
comparisons and gives more weight to larger trials and less
weight to smaller trials. This meta-analysis did not use the
results from an intention-to-treatment (ITT) approach;
alternatively, a per-protocol approach was used in calcu-
lating the pooled effect estimates for the combination of
single effects of trials. The fixed and random effect models
were applied to perform meta-analysis in the specific
regions of interest (ROIs) (femoral neck, trochanter, and
lumbar spine). When the study included two exercise
groups, the number of control group subjects was divided
among comparisons to ensure that the control group par-
ticipants were counted only once in the meta-analysis.
Heterogeneity of results between studies was deter-
mined by Q as well as I2. Statistical significance for Q was
set at an alpha value of B0.10. For I2, values of 25 to
\50 % were considered low heterogeneity, 50 to \75 %
moderate heterogeneity, and C75 % high heterogeneity.
Testing for overall effect (Z score) was regarded as sig-
nificant at p \ 0.05.
3 Results
Six controlled trials or RCTs including a total of 377
premenopausal women met the predetermined criteria [11–
16]. The flow diagram for the review process, including the
exclusion criteria, is summarized in Fig. 1, and general
descriptions of the characteristics of each study are inclu-
ded in Table 2.
3.1 Study Characteristics
For the analyzed studies, the total number of exercise
groups was greater than the number of control groups
because one study included more than one exercise group
[16]. All of the studies were conducted with healthy pre-
menopausal women between the ages of 20.5 ± 0.6 and
41.3 ± 3.8 years (mean ± SD). The sample sizes varied
from 27 to 120 participants. Premenopausal status was
determined by menstrual history and self-completed life-
style questionnaires. Two studies were conducted in the
UK [11, 12], two in Japan [13, 14], one in Finland [15] and
one in the US [16].
Dropout data were reported in both the exercise and
control groups [13–16]. There were some studies that had
withdrawals, but the relevant information was unclear [11,
12]. The reasons for dropping out of or being excluded
from exercise groups included changed circumstances,
time constraints, pain, personal issues, pregnancy, meno-
pause or irregular menstrual cycles, moving, loss of interest
and non-compliance with the exercise intervention. For the
control groups, reasons included changed circumstances,
123
1396
Fig. 1 Flow diagram for the selection of studies
time constraints, personal issues, pregnancy, menopause or
irregular menstrual cycles, moving, and loss of interest.
3.2 Study Quality Assessment
The quality assessment instrument scores awarded to studies
ranged from 1 to 3. With the exception of one study that
received a quality score of three [15], five studies received a
quality score of one or two [11–14, 16]. Five of the studies
were awarded one point for randomization, and one study was
allocated an extra point for describing the randomization
process [15]. All studies were awarded one point for state-
ments regarding participant withdrawal. No study gained
points for blinding of participants or contained a description of
adequate concealment of allocation to study or control groups.
3.3 Exercise Intervention
All studies compared the efficacy of a jumping exercise
intervention with respective control groups. A description
of training program characteristics is shown in Table 2.
Briefly, all studies prescribed high-impact exercise
(jumping) with one exercise plus lower or lower/upper
body resistance exercise [16] and one exercise combining
stamping, running, and walking [15]. Study duration ran-
ged from 6 to 12 months [11–16], with a training frequency
of two to three times per week in most interventions. Two
exercise protocols were conducted six or seven times per
week, but the intensity of training was relatively low [11,
12], and the total training volume was similar to other
studies. Three control groups were asked to maintain their
usual diet and lifestyle [12, 13, 15]. Two control groups
performed mild arm exercises or stretching and balance
123
R. Zhao et al.
exercises [11, 14]. Specific information was not available
on one of the control groups.
3.4 Outcome Measurement
The primary bone health outcome for all studies was BMD
on DXA. A description of BMD measurement in the ROIs
is provided in Table 2. Specific ROIs were assessed by
DXA in the femoral neck, trochanter and lumbar spine.
Absolute change values in BMD, along with SDs, after the
exercise interventions were available for two studies [11,
12]. Absolute change values, along with SDs, after the
interventions were calculated from the initial and final
measurements were available for four studies [13–16].
3.5 Meta-Analysis
The included trials represented data from 377 premeno-
pausal women included in six studies. A total of 201
healthy premenopausal women performed a jumping
exercise intervention from 6 to 12 months in duration, and
176 participants were assigned to non-exercise control
groups. We first determined heterogeneity in the initial
meta-analysis including all study groups. The results
demonstrated that the level of heterogeneity was low in the
femoral neck (Q = 2.63, p = 0.854, I2 = 0.0 %), tro-
chanter (Q = 2.10, p = 0.10, I2 = 0.0 %), and lumbar
spine BMD (Q = 1.17, p = 0.979, I2 = 0.0 %). Accord-
ingly, fixed-effect models were applied for further meta-
analysis.
Jumping exercise significantly increased BMD in the
femoral neck (WMD [fixed effect] = 0.017 g/cm2, 95 % CI
0.014–0.20, Z = 11.01, p \ 0.001) [Fig. 2] and trochanter
(WMD [fixed effect] = 0.021, 95 % CI 0.018–0.024,
Z = 13.08, p \ 0.001) [Fig. 3]. However, the lumbar spine
seemed to obtain less benefit from such high-impact exercise
(WMD [fixed effect] = -0.001, 95 % CI -0.003 to 0.001,
Z = 1.34, p = 0.181). Funnel plots were produced for the
effect of jumping exercise interventions on BMD in the
femoral neck (Fig. 4) and trochanter (Fig. 5) for all analyzed
trials. Visual inspection of the plots implicated some degree
of asymmetry, indicating a positive treatment effect at the
femoral neck and trochanter sites.
4 Discussion
The purpose of this meta-analysis was to examine the
effect of jumping exercise on changes in BMD at the
femoral neck, trochanter and lumbar spine among pre-
menopausal women. The studies we analyzed were six
controlled trials or RCTs published in peer-reviewed lit-
erature. The outcomes for the influence of a jumping
 Table 2 General characteristics of included studies
Study Subject age Country/ Exercise intervention group Control group Length of Quality ROIs
(years; sample intervention assessment
mean ± SD) size (n) (months) score

Bassey and T: 32 ± 1.18 UK Weekly jumping and skipping modes of movement accounting for 5–10 % of Mild arm exercises 6 2 Femoral
Ramsdale C: 29.8 ± 1.8 T: 14 the total 1 h class time and daily home exercises consisting of 50 jumps at home neck
[11] Lumbar
C: 13
spine
Trochanter
Bassey et al. T: 38.4 ± 7.4 UK The intervention consisted of a few minutes of warm up, five bouts of 10 Usual diet and 6 2 Femoral
[12] C: 36.4 ± 7.6 T: 30 vertical jumps, and finished with stretching exercise, 6 days/week lifestyle neck
C: 25 Lumbar
spine
Trochanter
Kato et al. T: 20.5 ± 0.6 Japan Two-legged maximum vertical jumps 10 times using arm swing in Asked to maintain 6 2 Femoral
[13] C: 20.9 ± 0.8 T: 21 countermovement style, 3 days per week previous lifestyle neck
C: 21 Lumbar
Jumping Exercise Impact on Bone Health in Postmenopausal Women

spine
Trochanter
Niu et al. T: 39.7±1.2 Japan Each exercise section included five bouts of 10 vertical and versatile jumps Received stretching 12 2 Femoral
[14] C: 38.1±1.2 T: 34 with two legs together and arms swinging in countermovement style, 3 times and balance neck
a week exercise Lumbar
C: 33
spine
Trochanter
Vainionpää T: 38.1 ± 1.7 Finland Exercise including 40-min high-impact activity (jumping, stamping, running, Normal daily life 12 3 Femoral
et al. [15] C: 38.5 ± 1.6 T: 60 and walking), 3 times a week and current neck
physical activity Lumbar
C: 60
spine
Trochanter
Winters- T1: USA Treatment 1: The lower body training program consisted of nine sets of 10–12 No specific 12 1 Femoral
Stone and 38.3 ± 3.8 T1: 21 jumps and nine sets of 10–12 repetitions of lower body resistance exercises information for neck
Snow [16] T2: Treatment 2: Following nine sets of 10–12 jumps and lower body resistance control group Lumbar
T2: 21
41.3 ± 3.8 exercise, participants performed three sets of 8–12 repetitions of upper body spine
C: 24
C: 40.5 ± 3.5 exercises, 3 times a week Trochanter
T treatment group, C control group, T1 or T2 group with type 1 or 2 exercise protocols, SD standard deviation, ROIs regions of interest
1397

123
1398 R. Zhao et al.

Fig. 2 Forest plot for changes in femoral neck BMD with jumping exercise. The dotted line represents the line of mean treatment effect. The
diamond denotes overall treatment effect with 95 % CI. WMD weighted mean difference, BMD bone mineral density, CI confidence interval

Fig. 3 Forest plot for changes in trochanter BMD with jumping exercise. The dotted line represents the line of mean treatment effect. The
diamond denotes overall treatment effect with 95 % CI. WMD weighted mean difference, BMD bone mineral density, CI confidence interval

exercise intervention on femoral neck and trochanter BMD
were significant, but the results in the lumbar spine were
inconsistent.
4.1 Femoral Neck and Trochanter
Our structured systematic searches resulted in six papers
and seven study group comparisons to evaluate the changes
in BMD at the femoral neck, trochanter, and lumbar spine
resulting from exercise interventions. The majority of the
RCTs included in this review reported absolute changes in
BMD. Jumping exercise proved to be effective in
improving bone health, bringing approximately
0.017 ± 0.474 and 0.021 ± 0.443 g/cm2 increases in
femoral neck and trochanter BMD, respectively, among
premenopausal women. Based on a cohort study,

123
Jumping Exercise Impact on Bone Health in Postmenopausal Women
Fig. 4 Funnel plot for femoral neck BMD outcomes from RCTs
including 95 % CI lines. The vertical line represents zero size (no
treatment effect). WMD weighted mean difference, BMD bone
mineral density, RCTs randomized controlled trials, CI confidence
interval
Fig. 5 Funnel plot for trochanter BMD outcomes from RCTs
including 95 % CI lines. The vertical line represents zero size (no
treatment effect). WMD weighted mean difference, BMD bone
mineral density, RCTs randomized controlled trials, CI confidence
interval
Cummings et al. [24] suggested that bone density of the
femoral neck was a better predictor than measurements of
the spine, radius, and calcaneus, and that each SD decrease
in femoral neck bone density increased the risk of hip
fracture 2.6 times. Therefore, the increment in femoral
neck and trochanter BMD were clinically significant
because the improvement in BMD effectively prevented
bone loss and could decrease the risk of hip fracture.
Combining this with the benefits resulting from exercise-
related muscle mass, strength gain and dynamic balance
could increase the overall effect in reducing fracture risk
[24].
1399
4.2 Lumbar Spine
Based on the meta-analysis, we found no significant effect
of jumping exercise on BMD in the lumbar spine. These
findings suggested that, unlike the femoral neck and tro-
chanter, the lumbar spine was less sensitive in its response
to high-impact exercise. It seems that the mechanical
strains generated against the ground during jumping exer-
cise may have been attenuated before transmission to the
spine and did not generate a sufficient osteogenic stimulus
for bone modeling [25]. Additionally, the responses of
bone to a high-impact mechanical stimulus may be site-
specific [26]. Martyn-St James and Carroll [27] reported
that high-impact-only protocols were effective only at the
femoral neck and that the optimal impact exercise pro-
grams to best preserve lumbar spine BMD were protocols
that combined impact exercise with resistance exercise.
Bone can adapt both materially and structurally to
increased mechanical loading. In all of the studies we
analyzed, the measurement of BMD was made with the aid
of DXA. However, BMD may not be the optimal means to
examine bone strength as it does not include another
important bone strength predictor, bone architecture.
Because even small changes in bone geometry and struc-
ture can significantly improve bone strength, the mea-
surement of structural changes is important in the
determination of exercise-related skeletal changes. Recent
data have suggested that BMD and the geometrical adap-
tion of bone to mechanical loading vary by age, skeletal
site, and sex [28]. Furthermore, evidence indicates that
older women can increase their bone strength by increasing
the structural characteristics of bone [29]. Therefore, BMD
measurement alone cannot determine whether the lumbar
spine benefits from jumping exercise.
4.3 Exercise Intervention
The exercise intervention employed in a majority of the
studies involved two-legged jumps with countermovement
style. This predominant exercise style gives more support
to the back and lower limbs with a lower risk of injury. In
addition, due to its brief and submaximal nature, this type
of high-impact exercise reduces the possibility of exercise-
related injuries. Two of the intervention groups involved
multidirectional jumps and movement, which created non-
habitual strains to the bone [14, 15]. Participants in all of
the studies were young to middle age adults with low risk
of developing osteoporosis and osteoporotic fractures.
Caution was still necessary for postmenopausal women or
women at risk of fracture during high-impact exercise.
Regular physical activity is recommended as an optimal
strategy for the prevention of osteoporosis. However, poor
rates of adherence to regular exercise beyond the study
123
1400
lifetimes represent a potential barrier to the improvement
of bone health in premenopausal women. Promoting the
integration of simple exercises into daily life will be of
great benefit to premenopausal women in preserving their
bone health [30, 31]. Except for one population-based
study [15], which may have affected exercise compliance,
the RCTs included in this review included home- or office-
based exercise requiring no equipment, travel and facility
cost, or time. The regimens in the studies promoted prac-
tical lifestyle interventions, which showed undoubted merit
in contrast to other studies [18, 26, 32, 33] because of their
minimal obligation and easier sustainability.
The intensity of the exercise interventions employed in
most studies in this review varied, but this did not result in
different outcomes. This suggests that the response of bone to
mechanical loading is not entirely intensity-dependent.
Studies have demonstrated that increases in loading cycles did
not proportionally produce osteogenic outcomes; adaptive
responses from 100 jumps a day was not significantly dif-
ferent from those of 40 jumps a day [34, 35]. This indicates
that extra mechanical loading exceeding a minimum thresh-
old will not necessarily result in more BMD gains. In addition
to workload, the interval between bouts of high-impact
exercise was also an important factor in stimulating bone
formation. Recent findings have demonstrated that by
allowing a short rest period between loading bouts, the oste-
ogenic effectiveness of subsequent cycles can be increased
[36, 37]. The insertion of short rest periods between loading
bouts can significantly increase intracellular calcium signal-
ing and enhance fluid flow near the osteocytes in successive
cycles [38–40]. It is now well-established that rest insertion
serves as a vital tool for ensuring that low-frequency, high-
impact exercise is a potent osteogenic stimulus. The interval
rest period between bouts of jumping exercise in this review
ranged from 8 to 15 s, which is consistent with the 15 s ele-
vation in intracellular calcium signaling that promoted bone
cell response to mechanical stimuli in animal studies [37, 38].
Therefore, it seems that the short rest period insertions
observed in this review may have augmented the efficiency of
jumping exercise.
4.4 Study Quality Assessment
Aspects of the methodological quality of the studies,
including randomization, blinding, and statements on
withdrawal, were assessed using a widely utilized instru-
ment [23]. In the present meta-analysis, the quality score of
included trials was relatively low. With the exception of
one study that received a quality score of three [15], five
studies received a quality score of one or two [11–14, 16].
None of the studies included a statement regarding whether
concealment of allocation occurred or not. Inadequate
concealment of allocation frequently results in an
123
R. Zhao et al.
overestimation of the effects of treatment and is generally
considered to be associated with bias.
4.5 Examination of Bias
In the current meta-analysis, we conducted a constructed
and systematic review to reduce an unclear risk of bias for
selective outcome reporting. Examination of funnel plots
suggested some asymmetry of positive effect size for
femoral neck and trochanter BMD outcomes. However, the
small number of study group comparisons available for
funnel plot interpretation was most likely not sufficient to
distinguish real asymmetry. Selective outcome reporting
may be considered as a subset of findings that are reported
based on their results [41]. Therefore, the major concern is
that results without statistical significance may be withheld.
As a result, the current meta-analysis most likely overes-
timated treatment effects. Additionally, the inability to
present detailed BMD outcomes prevented the investiga-
tive team from retrieving necessary data from original
papers and required us to exclude some eligible studies [17,
18]. Because of this, this potential form of bias cannot be
excluded from the current study. With the exception of one
trial [15], most of the included studies did not present valid
ITT data, so potential bias as a result of attrition cannot be
accounted for. Other factors also contribute to the risk of
bias, including study quality, choice of precision measures
and choice of effect measures. Given these questions in our
meta-analysis, potential bias cannot be ruled out, and we
suggest that future studies provide more detailed informa-
tion, such as original registry number and proper BMD
outcomes, so that potential bias can be determined.
4.6 Limitation
The main limitations of this review were the inclusion of
only six studies and the highly selective samples of pre-
menopausal women of varying ages. Future studies should
include both traditional resistance exercises and high-
impact intervention studies to examine the optimal exercise
for improving premenopausal bone health.
5 Conclusion
This meta-analysis indicated that skeletal response to
jumping exercise was site-specific, with high sensitivity at
the femoral neck and trochanter and low sensitivity at the
lumbar spine. In addition to mechanical loading, the
insertion of short rest periods between loading bouts during
jumping exercise may augment the response of bone to
mechanical stimuli in premenopausal women. Practice of
home- or office-based, short duration, low-frequency, high-
Jumping Exercise Impact on Bone Health in Postmenopausal Women
impact jumping exercise may be a feasible strategy for
improving bone health in premenopausal women.
There remains a need for well-designed, long-term
RCTs with an adequate sample size to quantify the effect of
jumping exercise on bone strength and its determinants.
Further detailed studies are needed to characterize both the
material and structural changes underpinning high-impact
exercise-induced gains in bone strength because even small
changes in bone geometry and structure can significantly
improve bone strength. Nevertheless, the high impact nat-
ure of jumping exercise coupled with its physiological
benefits proposes a practical non-pharmacological strategy
for premenopausal bone loss.
Acknowledgments This article was based on work funded by the
Zhejiang Provincial Natural Science Foundation of China under grant
numbers Y2110954 and LY14H070001. Renqing Zhao, Meihua
Zhao, and Liuji Zhang have no potential conflicts of interest that are
directly relevant to the content of this article.
References
1. World Health Organization. Assessment of fracture risk and its
application to screening for postmenopausal osteoporosis. WHO
Technical Report Series. Geneva: World Health Organization;
1994.
2. National Osteoporosis Foundation. America’s bone health: the
state of osteoporosis and low bone mass in our nation. Wash-
ington, DC: National Osteoporosis Foundation; 2002.
3. Zhao D, Wu H, Liu Z. Epidemiology of osteoporosis. In: Liu Z,
editor. Bone mineral and clinic practise. Beijing: Science and
Technology Press; 2006. p. 5–6.
4. Riggs BL, Melton LJ 3rd. The worldwide problem of osteopo-
rosis: insights afforded by epidemiology. Bone. 1995;17(5 Sup-
pl):505S–11S.
5. Vondracek SF, Hansen LB, McDermott MT. Osteoporosis risk in
premenopausal women. Pharmacotherapy. 2009;29(3):305–17.
6. Kohrt WM, Bloomfield SA, Little KD, et al. American College of
Sports Medicine Position Stand: physical activity and bone
health. Med Sci Sports Exerc. 2004;36(11):1985–96.
7. Wallace BA, Cumming RG. Systematic review of randomized
trials of the effect of exercise on bone mass in pre- and post-
menopausal women. Calcif Tissue Int. 2000;67(1):10–8.
8. Wolff I, van Croonenborg JJ, Kemper HC, et al. The effect of
exercise training programs on bone mass: a meta-analysis of
published controlled trials in pre- and postmenopausal women.
Osteoporos Int. 1999;9(1):1–12.
9. Chilibeck PD, Sale DG, Webber CE. Exercise and bone mineral
density. Sports Med. 1995;19(2):103–22.
10. Forwood MR, Larsen JA. Exercise recommendations for osteo-
porosis. A position statement of the Australian and New Zealand
Bone and Mineral Society. Aust Fam Physician. 2000;29(8):
761–4.
11. Bassey EJ, Ramsdale SJ. Increase in femoral bone density in
young women following high-impact exercise. Osteoporos Int.
1994;4(2):72–5.
12. Bassey EJ, Rothwell MC, Littlewood JJ, et al. Pre- and post-
menopausal women have different bone mineral density respon-
ses to the same high-impact exercise. J Bone Miner Res.
1998;13(12):1805–13.
1401
13. Kato T, Terashima T, Yamashita T, et al. Effect of low-repetition
jump training on bone mineral density in young women. J Appl
Physiol. 2006;100(3):839–43.
14. Niu K, Ahola R, Guo H, et al. Effect of office-based brief high-
impact exercise on bone mineral density in healthy premeno-
pausal women: the Sendai Bone Health Concept Study. J Bone
Miner Metab. 2010;28(5):568–77.
15. Vainionpää A, Korpelainen R, Leppaluoto J, et al. Effects of
high-impact exercise on bone mineral density: a randomized
controlled trial in premenopausal women. Osteoporos Int.
2005;16(2):191–7.
16. Winters-Stone KM, Snow CM. Site-specific response of bone to
exercise in premenopausal women. Bone. 2006;39(6):1203–9.
17. Bailey CA, Brooke-Wavell K. Optimum frequency of exercise
for bone health: randomised controlled trial of a high-impact
unilateral intervention. Bone. 2010;46(4):1043–9.
18. Heinonen A, Kannus P, Sievanen H, et al. Randomised controlled
trial of effect of high-impact exercise on selected risk factors for
osteoporotic fractures. Lancet. 1996;348(9038):1343–7.
19. Sugiyama T, Yamaguchi A, Kawai S. Effects of skeletal loading
on bone mass and compensation mechanism in bone: a new
insight into the ‘‘mechanostat’’ theory. J Bone Miner Metab.
2002;20(4):196–200.
20. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items
for systematic reviews and meta-analyses: the PRISMA state-
ment. Ann Intern Med. 2009;151(4):264–9.
21. Manolagas SC. Birth and death of bone cells: basic regulatory
mechanisms and implications for the pathogenesis and treatment
of osteoporosis. Endocr Rev. 2000;21(2):115–37.
22. Parfitt AM. Osteonal and hemi-osteonal remodeling: the spatial
and temporal framework for signal traffic in adult human bone.
J Cell Biochem. 1994;55(3):273–86.
23. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of
reports of randomized clinical trials: is blinding necessary?
Control Clin Trials. 1996;17(1):1–12.
24. Cummings SR, Black DM, Nevitt MC, et al. Bone density at var-
ious sites for prediction of hip fractures. The Study of Osteoporotic
Fractures Research Group. Lancet. 1993;341(8837):72–5.
25. O’Coventry E, Connor KM, Hart BA, et al. The effect of lower
extremity fatigue on shock attenuation during single-leg landing.
Clin Biomech (Bristol, Avon). 2006;21(10):1090–7.
26. Vainionpää A, Korpelainen R, Sievanen H, et al. Effect of impact
exercise and its intensity on bone geometry at weight-bearing
tibia and femur. Bone. 2007;40(3):604–11.
27. Martyn-St James M, Carroll S. Effects of different impact exer-
cise modalities on bone mineral density in premenopausal
women: a meta-analysis. J Bone Miner Metab. 2010;28(3):
251–67.
28. Nikander R, Sievanen H, Heinonen A, et al. Targeted exercise
against osteoporosis: a systematic review and meta-analysis for
optimising bone strength throughout life. BMC Med. 2010;8:47.
29. Ahlborg HG, Johnell O, Turner CH, et al. Bone loss and bone size
after menopause. N Engl J Med. 2003;349(4):327–34.
30. Blair SN, Kohl HW, Gordon NF, et al. How much physical activity
is good for health? Annu Rev Public Health. 1992;13:99–126.
31. Winters KM, Snow CM. Detraining reverses positive effects of
exercise on the musculoskeletal system in premenopausal
women. J Bone Miner Res. 2000;15(12):2495–503.
32. Kontulainen S, Heinonen A, Kannus P, et al. Former exercisers of
an 18-month intervention display residual aBMD benefits com-
pared with control women 3.5 years post-intervention: a follow-
up of a randomized controlled high-impact trial. Osteoporos Int.
2004;15(3):248–51.
33. Friedlander AL, Genant HK, Sadowsky S, et al. A two-year
program of aerobics and weight training enhances bone mineral
density of young women. J Bone Miner Res. 1995;10(4):574–85.
123
1402
34. Rubin CT, Lanyon LE. Regulation of bone formation by applied
dynamic loads. J Bone Joint Surg Am. 1984;66(3):397–402.
35. Umemura Y, Ishiko T, Yamauchi T, et al. Five jumps per day
increase bone mass and breaking force in rats. J Bone Miner Res.
1997;12(9):1480–5.
36. Robling AG, Burr DB, Turner CH. Partitioning a daily
mechanical stimulus into discrete loading bouts improves the
osteogenic response to loading. J Bone Miner Res. 2000;15(8):
1596–602.
37. Srinivasan S, Weimer DA, Agans SC, et al. Low-magnitude
mechanical loading becomes osteogenic when rest is inserted
between each load cycle. J Bone Miner Res. 2002;17(9):1613–20.
38. Donahue SW, Donahue HJ, Jacobs CR. Osteoblastic cells have
refractory periods for fluid-flow-induced intracellular calcium
123
R. Zhao et al.
oscillations for short bouts of flow and display multiple low-
magnitude oscillations during long-term flow. J Biomech.
2003;36(1):35–43.
39. Donahue SW, Jacobs CR, Donahue HJ. Flow-induced calcium
oscillations in rat osteoblasts are age, loading frequency, and
shear stress dependent. Am J Physiol Cell Physiol.
2001;281(5):C1635–41.
40. Lanyon LE. Functional strain in bone tissue as an objective, and
controlling stimulus for adaptive bone remodelling. J Biomech.
1987;20(11–12):1083–93.
41. Hahn S, Williamson PR, Hutton JL, et al. Assessing the potential
for bias in meta-analysis due to selective reporting of subgroup
analyses within studies. Stat Med. 2000;19(24):3325–36.