Skip Navigation LinksHome > March 2014 - Volume 46 - Issue 3 > Exercise to Improve Pediatric Bone and Fat: A Systematic Re...
Medicine & Science in Sports & Exercise:
doi: 10.1249/MSS.0b013e3182a6ab0d
Applied Sciences

Exercise to Improve Pediatric Bone and Fat: A Systematic Review and Meta-analysis


Free Access
Article Outline
Collapse Box

Author Information

1Centre for Musculoskeletal Research, Griffith Health Institute, Gold Coast, Queensland, AUSTRALIA; 2School of Allied Health Sciences, Griffith University, Gold Coast, Queensland, AUSTRALIA

Address for correspondence: Belinda Beck, Ph.D., F.A.C.S.M., School of Allied Health Sciences, Griffith University Gold Coast campus, QLD 4222, Australia; E-mail:

Submitted for publication April 2013.

Accepted for publication July 2014.

Collapse Box


Purpose: This study aimed to determine the effects of school-based, bone-focused exercise interventions on bone, fat, and lean mass in children by systematically reviewing and meta-analyzing the literature.

Methods: Potentially relevant articles were identified by searching electronic databases. Abstracts were included if they described the effects of an in-school exercise intervention for children 5–17 yr old compared with controls and presented baseline and follow-up results for bone, fat, and lean measures. Identified studies were systematically reviewed for methodological quality. Meta-analyses were performed for whole body, lumbar spine, and femoral neck bone mineral content (BMC), fat, and lean mass.

Results: Sixteen eligible trials were identified including eight randomized controlled trials, three clinical controlled trials, and five nonrandomized, nonmatched studies. The quality analysis revealed two studies had low, nine had medium, and five had a high risk of bias. Meta-analyses revealed a small positive effect of bone-targeted exercise on whole body BMC (standardized mean difference [SMD] = 0.483, 95% CI = 0.132–0.833), femoral neck BMC (SMD = 0.292, 95% CI = −0.022 to 0.607), lumbar spine BMC (SMD = 0.384, 95% CI = 0.193–0.575), fat mass (SMD = −0.248, 95% CI = −0.406 to −0.089), and lean mass (SMD = 0.159, 95% CI = −0.076 to 0.394).

Conclusions: Beneficial effects of school-based, bone-targeted exercise were observed for bone and fat, but not for lean mass. Excluding trials with high risk of bias strengthened that effect. Considerable study heterogeneity may have obscured effects on lean mass. The effects observed for bone and fat support the pursuit of brief, jumping-focused interventions to reduce fat as well as enhance musculoskeletal tissue in school age children.

A lack of physical activity and inappropriate nutrition in youth is associated with the development of obesity and lifetime chronic disease (3,41,58). It is estimated that 35% of the global population is overweight—a proportion that is growing, particularly among children (2). Overweight and obesity during childhood are significant predictors of obesity in adulthood (7,10,18,29,49,53). More than 10% of the world’s school-age children are overweight, signaling a burgeoning public health problem (22,40). Both developed and developing countries have increased prevalence, with up to 30% of children overweight in the United States, China, United Kingdom, and Australia (11,42). Physical inactivity is also associated with deleterious effects on lean body mass elements, including bone and muscle (44). In 2000, there were 9 million osteoporotic fractures globally (19), and the incidence continues to rise due to an aging population. Although osteoporosis is a condition that typically occurs in the elderly, physical activity during childhood provides a critical stimulus to the growing musculoskeletal system that may prevent osteoporotic fracture in old age (32,46,54,56).

It is well known that physical activity is an effective method of reducing overweight, optimizing cardiovascular, musculoskeletal, and metabolic health and preventing chronic diseases (39,50). It is also well recognized that the U.S. National Health Survey physical activity recommendations are rarely met (3) and that physical inactivity persists into adulthood (49). The challenge of reaching daily physical activity targets is increased by differences in the nature, intensity, and duration of exercises recommended for different tissues or systems. For example, non–weight-bearing aerobic exercise that may be effective for fat loss will not adequately stimulate the skeleton to promote gains in bone. Performing multiple workouts for different body systems is highly impractical and decreases the likelihood that exercise recommendations will be met. In fact, it would be of great practical utility to determine whether physical activity interventions that are known to enhance bone also reduce fat. Some physical activity interventions have provided such evidence (51,57); however, those data are rarely reported together.

Thus, the aim of the current work was to systematically review the findings of school-based physical activity interventions that were designed to improve bone health, to determine whether improvements in muscle and reductions in fat were also observed. A finding that broader health benefits for body composition can be obtained from bone-focused workouts for school-age children will inform recommendations for pediatric exercise interventions.

Back to Top | Article Outline


Back to Top | Article Outline
Eligibility criteria

Comprehensive database searches were conducted in the first instance for the following predefined keywords: (1) “children,” “adolescents,” “kids,” “paediatrics,” and “pediatrics” (yielded 2,408,264 studies); (2) “exercise,” “physical activity,” “fitness,” and “energy expenditure” (yielded 707,741 studies); (3) “obesity,” “obese,” “overweight,” “body fat,” and “body weight” (yielded 658,708 studies); and (4) “bone,” “bone mass,” “bone density,” and “bone strength” (yielded 1,309,830 studies). As a second step, the four categories were combined to obtain studies including all of the search terms. Finally, all of the potential studies were manually analyzed for inclusion. The following search criteria were additionally applied: humans, male, female, meta-analysis, randomized controlled trial, review, journal article, English, Portuguese, and child (0–18 yr). In an attempt to acquire all relevant literature, the search was conducted in 12 databases: Medline (OvidSP), PubMed (NLM), CINAHL (EBSCO), Cochrane Central Register of Controlled Trials (CENTRAL), Embase (OvidSP), PreMedline (OvidSP), ProQuest, PsycInfo (OvidSP), SPORTDiscuss (EBSCO), LILACS (Latin American and Caribbean Health Sciences), Web of Science and PEDro, up to August 2012. Where conference abstracts or reports were found through Web searches, we attempted to make contact with the investigators to identify studies in progress or unpublished data. If the required data (e.g., mean group change or SD) were not presented in reviewed publications, authors were contacted by email to request additional data (17,33). Dissertations and abstracts were not included in the review. Finally, reference lists of relevant articles were hand searched.

Back to Top | Article Outline
Selection of the studies

The studies were defined as eligible if they described controlled, but not necessarily randomized, school-based physical activity interventions for healthy children between 5 and 17 yrs of age, inclusive. The trials must have implemented at least one type of exercise intervention, included baseline and follow-up testing, and reported outcomes for bone and fat mass. The control group must have continued with normal physical education classes, according to the school curriculum. Identified articles and abstracts were screened and assessed for methodological quality based on titles and abstract. If eligibility was uncertain, the full text was consulted. A PRISMA flow diagram summarizing the process of identification, screening, and eligibility of articles is presented in Figure 1 (37).

Figure 1
Figure 1
Image Tools
Back to Top | Article Outline
Data collection, analysis, and quality assessment

A single reviewer (RN) assessed and screened all manuscripts. In cases of uncertainty, a second reviewer was consulted. After duplicate articles were removed from the search, remaining abstracts of each study were re-reviewed to verify content. Data extraction was performed by a single reviewer (RN). Data extracted included population, age, baseline characteristics, intervention activities, and outcome measures.

Studies were ranked by quality according to degree of bias risk. Because it is not possible to blind exercise intervention trials, it was not applied as a criterion of validity. The grading system, based on a previously published system (17), is presented in Table 1. The highest possible quality grade was 21. For the purpose of the current review, a grade of 19–21 indicated low risk of bias, 16–18 indicated moderate risk, and a grade of 15 or below indicated high risk of bias.

Table 1
Table 1
Image Tools
Back to Top | Article Outline

After systematically reviewing for study quality, a meta-analysis was conducted to quantify outcomes of the review. Meta-analyses were performed for whole body (WB), lumbar spine (LS), and femoral neck (FN) bone mineral content (BMC), lean, and fat mass. Investigators for one study reported data for boys and girls independently, so those data were included in the meta-analysis separately (23). Not all studies that were systematically reviewed were included in the meta-analysis, as some did not present means and/or SD for cases and controls or change in outcome measures (4,6,12,47). One trial only presented data graphically (13). Meta-analyses were initially performed including only trials with low to medium risk of bias to draw conclusions from the most valid data. Data were then re-analyzed, including the high risk of bias trials to determine whether trial quality influenced outcomes.

Meta-analyses were carried out using the Meta-Analyst Beta software version 3.13 (Tufts Evidence-based Practice Center, Tufts Medical Center, Boston, MA) (55). A random-effects model was adopted. Study heterogeneity was examined using I2 as a measure of consistency, representing the percentage of total variation across studies for each meta-analysis. A value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity (15). Effect size was represented by the standardized mean difference (SMD) along with 95% confidence intervals, which was calculated by subtracting the mean of one group from the other (M1–M2) and dividing the result by the pooled SD of the population from which the groups were sampled. Weighted averages were calculated based on the study sample sizes and were automatically generated by the software. Variances of the intervention and control groups were generally similar, and thus, no further action was taken to weight effect sizes on this basis. The convention of Cohen (8) was used for describing the strength of the effect size, based on Cohen’s d. That is, 0.20 is a small effect, 0.50 is a medium effect, and 0.80 is a large effect (15). Meta-regressions were not undertaken owing to low study numbers (16).

Back to Top | Article Outline


Back to Top | Article Outline
Search results

A total of 1067 articles were identified. Of those, 832 articles were excluded based on the screening of titles and abstracts for duplicate publications, nonintervention study designs, or an absence of follow-up. Further exclusions were made from the remaining 235 articles if a study cohort included obese children, adults, nonhealthy children, and elite athletes, or if the results for both bone and fat mass were not presented (Fig. 1). Few articles reported both bone and fat outcome measures; thus, only 20 relevant articles could ultimately be included in the systematic review (4,6,12,13,20,21,23,26–28,33–35,38,47,51,52,57,59,60). Notably, some authors reported the same study in multiple publications, typically reporting different intervention intervals (20,51) or study cohorts (4,13,26,28,59,60). Of the twenty potentially relevant trials, four were excluded (27,51,59,60) because their data had been previously published in five other articles identified in our search (13,20,26,28,33). Therefore, the current review was based on the analysis of 16 trials. Reviewed trials were conducted by American, Australian, Belgian, British, Canadian, Swedish and Swiss research groups.

Back to Top | Article Outline
Study characteristics

The studies were heterogeneous in terms of study design, quality, target population, and outcome measures. An overview of the design, execution, and outcomes of each study is presented in Table 2. Of the 16 studies, 8 were randomized controlled trials (12,13,21,23,26,28,34,57), 3 were clinical controlled trials (20,33,52), and the remaining 5 were neither randomized nor well matched (4,6,35,38,47). The type, duration, and frequency of each exercise program varied widely, as did the intervention period, which ranged from 9 wk (35) to 3 yr (13). The frequency of exercise bouts varied from 2 to 5 d·wk−1, including one protocol that divided physical activity (PA) into three daily sessions, 5 d·wk−1 (33). The physical activity programs of seven studies were composed primarily of jumping activities (6,12,26,28,33,34,57), whereas others included aerobic exercise (4,13,20,21,23,35,38,47,52), stretching (4,13), or strength training (47).

Table 2
Table 2
Image Tools
Back to Top | Article Outline
Quality assessment

Articles were grouped according to their quality ranking (Table 2). Trial quality rankings ranged from 12 to 20, from highest to lowest risk of bias, respectively. Most of the trials were considered to have a moderate or high risk of bias based on a lack of randomization (6,20,21,33,35,38,47,52), unspecified or low study compliance (4,13,23,26,28,52) and/or intervention compliance (4,5,13,26,33,34), lack of ground reaction force (GRF) data (4,6,13,20,21,34,35,38,47,52,57) or a specific intervention program (4,13,20,21,34,35,38,47), short intervention period (6,35), and small sample size (6,35) and lack of control of variables, such as physical activity history and/or calcium consumption (4,6,13,20,21,23,26,28,35,52).

Back to Top | Article Outline
Outcome measures

Results are presented as the difference in percent change between intervention and control groups, including the P value for between-group differences if provided by the authors. Bone outcomes were primarily BMC (4,6,12,20,21,23,26,28,33,35,38,47,52,57), bone mineral density (BMD) (4,12,13,20,34,35,38,47,52,57), and volumetric BMD (20,21). Some trials also included bone area (26,28,33,57). Lean and fat tissue outcomes included absolute fat mass, absolute lean mass (LM), or fat-free soft tissue (4,6,13,20,21,23,26,28,33–35,38,47,52,57). Some studies also provided percent body fat (%BF) (4,6,12,13,35,48). Several articles calculated BMI, so despite the recognized limitations of the index (9,31), BMI was also included. BMC, BMD, lean, and fat mass were assessed using dual-energy x-ray absorptiometry in all studies reporting those parameters. In addition, some trials evaluated percent body fat using caliper-based anthropometry (12,35,52). One study also examined indices of bone quality using quantitative ultrasound, measuring speed of sound (SOS), broadband ultrasound attenuation, and stiffness index (SI) (57). Magnetic resonance imaging was used to analyze bone marrow adipose tissue volume and total body composition in two studies (4,6) (Table 1).

Back to Top | Article Outline
Exercise interventions

The trials included in the review were designed to primarily examine the effects of high-impact weight-bearing exercise on bone; however, some tested the effect of multiple exercise modes. Of the 16 trials, 6 used jump activities only, lasting an average of 10 min (12,26,28,33,34,57). One trial added a 5-min warm-up and a 5-min cool down to a 20-min jumping regime (6). One 10-min intervention was primarily composed of jumps but also included two nonimpact exercises (52). The remaining eight studies combined different types of exercise, such as strength training, aerobic training, stretching, skills development, and high-impact maneuvers for 30–80 min, two to five times per week (4,13,20,21,23,35,38,47). Control groups were generally asked to continue usual physical education classes. Some protocols implemented stretching activities for those students not involved in the exercise intervention (12,57).

Back to Top | Article Outline
Sample characteristics

Sample sizes ranged from 20 to 447. All but two trials (6,13) evaluated maturity using Tanner staging, age of peak height velocity, or bone age. The latter two methods are less common but considered to be more valid for determining biological maturity than the Tanner assessment (30,36). According to participant age and/or Tanner classification, 10 trials were conducted in prepubertal children (6,12,13,20,21,23,28,34,38,52) or early pubertal children (4), three in prepubertal and early pubertal children (26,33,35) and two in peripubertal children (47,57). Pubertal designation was assigned according to the original assignation of each report. The mean age of participants at baseline varied from 6 to 15 yr. The trials included black, white, Caucasian, and Asian boys and girls. Of the 16 studies analyzed, 8 were conducted in both sexes (6,12,13,23,33–35,57), while 2 were conducted with boys only (21,28).

Back to Top | Article Outline
Exercise effects on bone, fat, and lean mass

Of 16 trials, 14 reported at least one positive effect from exercise on bone mass in children and adolescents. Between-group effects varied from 1.1% (4) to 10.3% (38,57). The results for fat and lean mass were more varied. Some trials reported gains in lean mass (13,28,38), reductions in fat mass accumulation (4), or both in the exercise group compared with the control group (35,38,57). Nevertheless, some exercise interventions did not induce any significant difference in modification of body composition between groups (6,12,20,21,23,26,33,34,47,52). Trials will be discussed in order of quality, from least to most risk of bias.

Back to Top | Article Outline
Low risk of bias trials

The highest scoring trial, with a quality score (QS) of 20 (reflecting the lowest risk of bias) was conducted by Fuchs et al. (12). Their 7-month weight-bearing intervention was conducted with prepubertal children. They observed 4.9% greater improvements in FN BMC, 2.3% more FN area, 3.6% more LS BMC, and 2% more aBMD in the intervention group than control. No differences were observed between groups in terms of change in lean or fat mass.

The second strongest trial, an 8-month jumping regime for adolescents, was conducted by Weeks et al. (57), with a QS of 19. Whole body, FN, and TR BMC of the intervention group improved 3.6%, 6.0% and 10.3% more than the control group, respectively. Intervention boys gained more lean mass (2.3%) and lost more fat (−3.5%) than the control group.

Back to Top | Article Outline
Moderate risk of bias trials

MacDonald et al. (23) (QS 18) conducted an 11-month weight-bearing intervention for prepubertal boys and girls. Intervention group boys increased WB BMC and LS BMC 2.1% and 2.9% more than control, respectively. However, there were no between-group differences in change in fat and lean mass for boys or girls. MacKelvie et al. (28) conducted a 20-month exercise intervention for prepubertal boys (QS 17) and girls (26) (QS 16) but reported the findings separately. Intervention group boys increased FN BMC 4.5% and lean mass 4.2% more than the control group (28), whereas intervention group girls increased LS BMC 3.7% and FN BMC 4.6% more than control group. Neither fat nor lean mass change was different between groups for girls (26).

MacKay et al. (34) (QS 17) implemented a similar 8 month trial of weight-bearing exercises for pre and early pubertal children. Trochanteric aBMD (1.2%) increased more in the intervention group than that in the control group; however, no other outcome measure reached significance, including fat and lean mass. The study posed moderate risk of bias due to low study compliance and absence of reporting potentially confounding data (e.g., physical activity and calcium intake).

In contrast to the studies using targeted weight-bearing interventions, Linden et al. (20) (QS 17) examined the effect of four additional 40-min moderate-to-vigorous physical activity classes per week versus one standard 60-min PE class for 2 yr. The following were increased in the intervention group: WB aBMD, 1.5%; LS aBMD, 1.6%; LS BMC, 7.0%; and lean mass, 3.8% more than the control group; however, fat mass also increased by 9.5% more than the control group. The authors reported results only for girls but conducted an identical program with boys (21), presenting results after 1 yr (QS 16). There were no differences in change in WB or FN bone, lean, or fat mass between groups (21).

Schneider et al. (47) (QS 17) examined the effect of an 8-month aerobic and strength training protocol plus 500 mg·d−1 of calcium for adolescents, repeated annually for 3 yr. They observed a 4.9% greater increase in thoracic spine BMC in the exercise group than that in the control group, but no difference in change in any other bone parameter, fat, or lean mass was found. The latter studies (20,21,47) had some limitations, such as lack of randomization, nonspecific intervention activities, and no GRF reported, and were therefore considered at moderate risk of bias.

Morris et al. (38) conducted a 10-month exercise intervention for prepubertal girls, comprising general activities and weight-bearing training (QS 16). The intervention group exhibited 5.5%, 5.5%, and 4.5% greater increases in WB, LS, and FN BMC than that in the control group, respectively, along with 3.4% greater gains in lean mass and 6.2% less gain in fat. The study was assigned a QS of 16 due to a lack of randomization, nonspecific intervention activities, and the absence of GRF data (38).

In a study of twin prepubertal girls, Van Langendonck et al. (52) examined the effect of a 9-month, bone-specific training regime followed by some nonimpact movements. Greater increases were observed in FN aBMD (1.6%) and proximal femur BMC (2.4%) and aBMD (1.4%) in the intervention group compared with the control group. Fat mass increased by 3.8% and lean mass by 0.7% in the intervention group more than that in the control group; also, intervention participants reduced percent fat by 2.1% more than the control group. The study (52) was assigned moderate risk of bias (QS 16) due to a lack of GRF data, maturity level (although measured, not presented), and randomization.

Back to Top | Article Outline
High risk of bias trials

McKay et al. (33) implemented an 8-month translational frequent jumping trial for prepubertal children. The intervention group increased trochanteric BMC by 2.7% and area by 2.0%, as well as proximal femur BMC by 2.0% and area by 1.3% more than the control group. The control group, however, increased WB BMC 1.3% more than the intervention group. There were no between-group differences in change in lean and fat mass. Those findings are considered to be at high risk of bias (QS 15) due to the lack of randomization and low compliance in the intervention group. The tendency for higher physical activity in the intervention than the control group at baseline may also have confounded results (33).

Only one study did not report any positive results for BMC, BMD, or bone area (6). The aim of that pilot trial was to determine the effect of a 10-wk bone-specific exercise intervention on bone marrow adipose tissue in prepubertal children. Although significant changes in bone measures were not observed, a significant decrease in bone marrow adipose tissue was reported. The study, however, was considered at high risk of bias with a QS of 14 due to short duration, very small sample size, lack of randomization, and absence of GRF data (6).

One research group incorporated general physical activities into school time for prepubertal and early pubertal children and reported different group findings in two articles (both QS 13). One of the studies included boys and girls (13), and the other study included only black girls (4). In both cases, intervention groups increased WB BMD by 3.6% (13) and 1.1% (4) more than the control groups. Greater fat gain was observed in the control group than that in the intervention group, with an increase in percent fat for the control group only (4). Although both trials were randomized, there was low engagement (study compliance) and low protocol compliance, and confounding variables were not controlled, resulting in high risk of bias (4,13).

McWhannell et al. (35) reported the effects of a 9-wk high-impact vigorous activity intervention for prepubertal and early pubertal children compared with a control group and a lifestyle activities group. The intervention group increased WB BMC by 5.8% and BMD by 1.5% more than the control group. No significant differences were found between the lifestyle group and either the control or the intervention groups. There were no between-group differences in fat change. The trial was graded at very high risk of bias (QS 12) due to the lack of control for confounding variables, inadequate intervention period, small sample size, and lack of randomization (35).

Back to Top | Article Outline

Forest plots illustrate the results of the random-effects meta-analyses (Figs. 2–4). Including only the seven trials with low and medium risk of bias (20,21,26,28,38,52,57), a small effect of exercise on WB BMC was observed (SMD = 0.483, 95% CI = 0.132−0.833, I2 = 0.76). Including the two high risk of bias trials in the secondary meta-analysis (33,35), the observed effect remained small and the SMD was lower (SMD = 0.393, 95% CI = 0.043−0.744, I2 = 0.81).

Figure 2
Figure 2
Image Tools
Figure 3
Figure 3
Image Tools
Figure 4
Figure 4
Image Tools

A small effect (SMD = 0.292, 95% CI = −0.022 to 0.607, I2 = 0.62) of exercise on FN BMC was revealed from meta-analysis of the low to medium risk of bias trials (20,21,26,28,38,52,57). Again, inclusion of the high risk of bias trials weakened the effect (SMD = 0.270, 95% CI = 0.013−0.527, I2 = 0.59) (33,35).

A small effect of exercise on LS BMC was found from the six studies of low and medium risk of bias with available data (SMD = 0.384, 95% CI = 0.193−0.575, I2 = 0) (20,26,28,38,52,57). A smaller effect was observed when studies of high risk of bias were included (SMD = 0.263, 95% CI = 0.078−0.447, I2 = 0.21) (33,35).

From the nine low and moderate risk of bias studies including lean mass data (20,21,23,26,28,34,38,52,57), there was a very small effect of exercise on lean mass (SMD = 0.159, 95% CI = −0.076 to 0.394, I2 = 0.70). When the other two high risk of bias studies were included, the effect was slightly lower (SMD = 0.127, 95% CI = −0.073 to 0.328, I2 = 0.66) (33,35). The meta-analysis of fat mass from the eight low or medium risk trials with available data yielded a small effect of exercise (SMD = −0.248, 95% CI = −0.406 to −0.089, I2 = 0.33) (20,21,23,26,28,34,52,57), which was not markedly different from that observed when high risk of bias trials were included (SMD = −0.279, 95% CI = −0.434 to −0.124, I2 = 0.43) (33,35).

Back to Top | Article Outline


The goal of the current work was to investigate whether a range of school-based interventions designed to improve children’s bone health also reduced fat. Findings suggest that indeed the short bouts of high-intensity weight-bearing activity that can positively affect growing bone can in some cases also improve lean and fat tissue. Although our focus has been on interventions that may be appropriate for improving bone and reducing fat, we note that observations of simultaneous increases in lean mass are highly beneficial for both bone and fat metabolism. That is, increased muscle loading of the skeleton can stimulate bone adaptation (45), and increased muscle mass raises metabolism thereby assisting weight loss (14,43).

Although considerable research has been dedicated to evaluating the effect of in-school exercise interventions on body composition, the effects on bone, muscle, and fat are typically reported under separate cover. Similarly, reviews of studies investigating the benefits of school-based physical activity typically report bone and soft tissue outcomes separately (14,17).

The trials included in the current systematic review represent a variety of predominantly bone-targeted exercise interventions, including jumping (12,26,28,33,52,57), jumping and aerobic exercise (23), jumping and circuit training (34,38), or general moderate-to-vigorous circuit training (4,6,13,20,35). As not all trials reported GRF, it is not possible to quantify all reported loading regimes. However, it is well recognized that the prevailing high-impact loading activities used (such as jumping) are likely to be optimally osteogenic as a function of the high bone strain magnitudes and rates of loading they engender. Considerable differences in frequency, duration, and overall program length contributed to the marked study heterogeneity observed in WB and lean meta-analyses (high I2 > 0.05). Nevertheless, the most effective interventions included short (10- to 12-min) sessions of jumping exercises, executed two to three times per week (26,28,57).

In general, the trials that combined jumping activities with some other moderate-to-vigorous physical activity resulted in the greatest fat and lean mass changes, whereas the trials that were limited to jumping activities were more likely to report positive results for bone mass, but not for other body composition parameters. Intervention-related increases in bone mass were found at the WB (20,23,26,35,38,57), FN (12,21,23,26,28,35,38,52,57), and LS (21,26,28,35,52,57). The effect on muscle was more variable with some positive findings (4,13,28,38,57), but not all reported significant benefits (21,23,26,35,52). High study heterogeneity likely accounts for the inconsistent effects for lean mass (I2 = 0.70 and 0.66). Most trials, however, observed a reduction or less gain of fat in intervention groups compared with control groups (20,21,23,26,28,35,38,52,57). Low study heterogeneity may have enhanced the ability to detect an effect for fat (I2 = 0.33 and 0.43).

The quality scoring scheme used to quantify bias in the included studies (17) was an important tool to identify limitations and confounding factors in trials of exercise for bone. Ultimately, the QS performed the role of final screening for our meta-analyses and enhanced the validity of our findings by excluding data that are biased by weaknesses in methodology. There was considerable variation in trial quality with only two articles considered to be at low risk of bias, nine articles at moderate risk of bias, and five articles at high risk of bias. Ultimately, excluding the trials with a high risk of bias only strengthened the findings of the meta-analyses, suggesting that weaknesses in trial design and compliance may limit the ability to detect important exercise effects.

Our findings that short bouts of bone-focused exercise can improve WB, FN, and LS BMC in pediatric subjects corroborates those of previous bone-specific reviews (17,25). Although bone-specific programs produced a small effect size in relation to each measurement, the effect size was greater when the high risk of bias studies were excluded. Funnel plots for WB, LS, and FN BMC did not exhibit asymmetry, suggesting no notable publication bias (data not shown). Notably, low study heterogeneity was observed for the LS, whereas moderate and high heterogeneity were observed for FN and WB, respectively. The less consistent treatment effects on the FN and WB might reflect the relatively wide variety of exercise types and doses included and the nature of site-specific osteogenic responses.

Our observation that bone-targeted exercise programs may simultaneously reduce fat in pediatric groups is a fortuitous, if unintentional, outcome with positive implications for pediatric exercise prescription. Our lean mass observations were somewhat equivocal and, overall, indicative of only a weak effect of bone-targeted exercise programs. The considerable variation in lean mass outcomes between trials is likely a function of intervention heterogeneity, reflected by high I2 values. It is possible that jump-focused exercise programs do not contain a sufficient degree of progression to maximize muscle adaptation. Training progression is likely to be beneficial to both muscle and bone and should therefore be tested further.

Back to Top | Article Outline

Not all reviewed trials were included in the meta-analyses owing to an absence of numerical data (e.g., SD of the change after intervention) (4,6,12,13,24,47). Although LS and fat mass meta-analyses exhibited low study heterogeneity, FN BMC, WB BMC, and lean mass meta-analyses exhibited moderate to high study heterogeneity, limiting the generalizability of those findings. Meta-regressions to control study level covariates could not be performed due to the small number of trials (16).

It is noteworthy that most of the included trials were conducted in prepubertal children such that more mature Tanner stages were relatively underrepresented. It is possible that the prepubertal years are an important time, not only for optimizing bone accrual but also for minimizing fat accumulation. It is unknown if our findings would have differed if more studies from later Tanner stages could be included.

Back to Top | Article Outline


The benefits of jumping exercise for bone are well established, to the extent that it is recommended for the prevention of osteoporosis. The aim of the current systematic review was to determine whether pediatric exercise interventions designed to improve bone also improve soft tissue elements of body composition that influence risk of other chronic diseases. We found that indeed bone-targeted exercise programs will improve BMC and reduce fat mass (or at least mitigate gain). The evidence for simultaneous lean mass enhancement is less convincing. Although the optimal dose and type of exercise to improve bone, muscle, and fat have yet to be established, our findings suggest that broader benefits are to be gained from what has traditionally been considered a system-specific exercise program for bone. At a time when system-specific exercise guidelines are proliferating (1,61), our findings are cause for optimism that efficiencies can be achieved in exercise prescription.

No external funding was secured for this study.

The authors have no conflicts of interest to disclose.

The authors have no financial relationships relevant to this article to disclose.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Albright A, Franz M, Hornsby G, et al. American College of Sports Medicine position stand. Exercise and type 2 diabetes. Med Sci Sports Exerc. 2000; 32 (7): 1345–60.

2. Alwan A. Global Status Report on Noncommunicable Diseases 2010. World Health Organization; 2011. p. 176.

3. Andersen LB, Harro M, Sardinha LB, et al. Physical activity and clustered cardiovascular risk in children: a cross-sectional study (The European Youth Heart Study). Lancet. 2006; 368 (9532): 299–304.

4. Barbeau P, Johnson MH, Howe CA, et al. Ten months of exercise improves general and visceral adiposity, bone, and fitness in black girls. Obesity. 2007; 15 (8): 2077–85.

5. Boreham CAG, McKay HA. Physical activity in childhood and bone health. Brit J Sport Med. 2011; 45 (11): 877–9.

6. Casazza K, Hanks L, Hidalgo B, Hu H, Affuso O. Short-term physical activity intervention decreases femoral bone marrow adipose tissue in young children: a pilot study. Bone. 2012; 50 (1): 23–7.

7. Clarke WR, Lauer RM. Does childhood obesity track into adulthood? Crit Rev Sci. 1993; 33 (4–5): 423–30.

8. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. p. 567.

9. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Brit Med J. 2000; 320 (7244): 1240.

10. Craigie AM, Lake AA, Kelly SA, Adamson AJ, Mathers JC. Tracking of obesity-related behaviours from childhood to adulthood: A systematic review. Maturitas. 2011; 70 (3): 266–84.

11. Flynn MAT, McNeil DA, Maloff B, et al. Reducing obesity and related chronic disease risk in children and youth: a synthesis of evidence with ‘best practice’ recommendations. Obes Rev. 2006; 7: 7–66.

12. Fuchs RK, Bauer JJ, Snow CM. Jumping improves hip and lumbar spine bone mass in prepubescent children: a randomized controlled trial. J Bone Miner Res. 2001; 16 (1): 148–56.

13. Gutin B, Yin Z, Johnson M, Barbeau P. Preliminary findings of the effect of a 3-year after-school physical activity intervention on fitness and body fat: the Medical College of Georgia Fitkid Project. Int J Pediatr Obes. 2008; 3 (S1): 3–9.

14. Harris KC, Kuramoto LK, Schulzer M, Retallack JE. Effect of school-based physical activity interventions on body mass index in children: a meta-analysis. Can Med Assoc J. 2009; 180 (7): 719–26.

15. Higgins J, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Brit Med J. 2003; 327 (7414): 557–60.

16. Higgins JP, Green S, Collaboration C. Cochrane Handbook for Systematic Reviews of Interventions. 5th ed. Wiley Online Library; 2008.

17. Hind K, Burrows M. Weight-bearing exercise and bone mineral accrual in children and adolescents: A review of controlled trials. Bone. 2007; 40 (1): 14–27.

18. James PT, Leach R, Kalamara E, Shayeghi M. The worldwide obesity epidemic. Obes Res. 2012; 9 (S4): 228S–33 S.

19. Johnell O, Kanis J. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006; 17 (12): 1726–33.

20. Linden C, Ahlborg HG, Besjakov J, Gardsell P, Karlsson MK. A school curriculum-based exercise program increases bone mineral accrual and bone size in prepubertal girls: two-year data from the pediatric osteoporosis prevention (POP) study. J Bone Miner Res. 2006; 21 (6): 829–35.

21. Linden C, Alwis G, Ahlborg H, et al. Exercise, bone mass and bone size in prepubertal boys: one - year data from the pediatric osteoporosis prevention study. Scand J Med Sci Spor. 2007; 17 (4): 340–7.

22. Lobstein T, Baur L, Uauy R. Obesity in children and young people: a crisis in public health. Obes Rev. 2004; 5 (S1): 4–104.

23. Macdonald HM, Kontulainen S, Petit M, et al. Does a novel school-based physical activity model benefit femoral neck bone strength in pre- and early pubertal children? Osteoporos Int. 2008; 19 (10): 1445–56.

24. Macdonald HM, Kontulainen SA, Mackelvie-O’Brien KJ, et al. Maturity- and sex-related changes in tibial bone geometry, strength and bone-muscle strength indices during growth: a 20-month pQCT study. Bone. 2005; 36 (6): 1003–11.

25. MacKelvie KJ, Khan KM, McKay HA. Is there a critical period for bone response to weight-bearing exercise in children and adolescents? A systematic review. Br J Sports Med. 2002; 36 (4): 250–7.

26. MacKelvie KJ, Khan KM, Petit MA, Janssen PA, McKay HA. A school-based exercise intervention elicits substantial bone health benefits: a 2-year randomized controlled trial in girls. Pediatrics. 2003; 112 (6): e447–52.

27. MacKelvie KJ, McKay HA, Petit MA, Moran O, Khan KM. Bone mineral response to a 7-month randomized controlled, school-based jumping intervention in 121 prepubertal boys: associations with ethnicity and body mass index. J Bone Miner Res. 2002; 17 (5): 834–44.

28. MacKelvie KJ, Petit MA, Khan KM, Beck TJ, McKay HA. Bone mass and structure are enhanced following a 2-year randomized controlled trial of exercise in prepubertal boys. Bone. 2004; 34 (4): 755–64.

29. Magarey AM, Daniels LA, Boulton TJ, Cockington RA. Predicting obesity in early adulthood from childhood and parental obesity. Int J Obesity. 2003; 27 (4): 505–13.

30. Mahmoodi S, Sharif BS, Chester EG, Owen JP, Lee R. Skeletal growth estimation using radiographic image processing and analysis. IEEE Trans Inf Technol Biomed. 2000; 4 (4): 292–7.

31. Maynard LM, Wisemandle W, Roche AF Childhood body composition in relation to body mass index. Pediatrics. 2001; 107 (2): 344–50.

32. McKay H, Smith E. Winning the battle against childhood physical inactivity: the key to bone strength? J Bone Miner Res. 2008; 23 (7): 980–5.

33. McKay HA, MacLean L, Petit M, et al. “Bounce at the bell”: a novel program of short bouts of exercise improves proximal femur bone mass in early pubertal children. Br J Sports Med. 2005; 39 (8): 521–6.

34. McKay HA, Petit MA, Schutz RW, et al. Augmented trochanteric bone mineral density after modified physical education classes: a randomized school-based exercise intervention study in prepubescent and early pubescent children. J Pediatr. 2000; 136 (2): 156–62.

35. McWhannell N, Henaghan JL, Foweather L, et al. The effect of a 9-week physical activity programme on bone and body composition of children aged 10–11 years: an exploratory trial. Int J Sports Med. 2008; 29 (12): 941–7.

36. Mirwald RL, Baxter-Jones AD, Bailey DA, Beunen GP. An assessment of maturity from anthropometric measurements. Med Sci Sports Exerc. 2002; 34 (4): 689–94.

37. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg. 2010; 8 (5): 336–41.

38. Morris FL, Naughton GA, Gibbs JL, Carlson JS, Wark JD. Prospective ten-month exercise intervention in premenarcheal girls: positive effects on bone and lean mass. J Bone Miner Res. 1997; 12 (9): 1453–62.

39. Mountjoy M, Andersen LB, Armstrong N, et al. International Olympic Committee consensus statement on the health and fitness of young people through physical activity and sport. Brit J Sports Med. 2011; 45 (11): 839–48.

40. Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007–2008. J Am Med Assoc. 2010; 303 (3): 242–9.

41. Pekkinen M, Viljakainen H, Saarnio E, Lamberg-Allardt C, Mäkitie O. Vitamin D is a major determinant of bone mineral density at school age. PloS one. 2012; 7 (7): e40090.

42. Popkin BM. Recent dynamics suggest selected countries catching up to US obesity. Am J Clin Nutr. 2010; 91 (1): 284S–8.

43. Pratley R, Nicklas B, Rubin M, et al. Strength training increases resting metabolic rate and norepinephrine levels in healthy 50-to 65-yr-old men. J Appl Physiol. 1994; 76 (1): 133–7.

44. Rittweger J, Felsenberg D. Recovery of muscle atrophy and bone loss from 90 days bed rest: results from a one-year follow-up. Bone. 2009; 44 (2): 214–24.

45. Rubin CT, Lanyon L. Regulation of bone formation by applied dynamic loads. J Bone Joint Surg Am. 1984; 66 (3): 397–402.

46. Scerpella T, Dowthwaite J, Rosenbaum P. Sustained skeletal benefit from childhood mechanical loading. Osteoporos Int. 2011; 22 (7): 2205–10.

47. Schneider M, Dunton GF, Bassin S, et al. Impact of a school-based physical activity intervention on fitness and bone in adolescent females. J Phys Act Health. 2007; 4 (1): 17–29.

48. Sherk VD, Bemben DA, Bemben MG, Anderson MA. Age and sex differences in tibia morphology in healthy adult Caucasians. Bone. 2012; 50 (6): 1324–31.

49. Telama R, Yang X, Viikari J, et al. Physical activity from childhood to adulthood: a 21-year tracking study. Am J Prev Med. 2005; 28 (3): 267–73.

50. Tobias JH, Steer CD, Mattocks CG, Riddoch C, Ness AR. Habitual levels of physical activity influence bone mass in 11-year-old children from the United Kingdom: findings from a large population-based cohort. J Bone Min Res. 2006; 22 (1): 101–9.

51. Valdimarsson O, Linden C, Johnell O, Gardsell P, Karlsson MK. Daily physical education in the school curriculum in prepubertal girls during 1 year is followed by an increase in bone mineral accrual and bone width—data from the prospective controlled Malmo pediatric osteoporosis prevention study. Calcif Tissue Int. 2006; 78 (2): 65–71.

52. Van Langendonck L, Claessens A, Vlietinck R, Derom C, Beunen G. Influence of weight-bearing exercises on bone acquisition in prepubertal monozygotic female twins: a randomized controlled prospective study. Calcif Tissue Int. 2003; 72 (6): 666–74.

53. Venn AJ, Thomson RJ, Schmidt MD, et al. Overweight and obesity from childhood to adulthood: a follow-up of participants in the 1985 Australian Schools Health and Fitness Survey. Med J Australia. 2007; 186 (9): 458–60.

54. Vicente-Rodríguez G. How does exercise affect bone development during growth? Sports Med. 2006; 36 (7): 561–9.

55. Wallace BC, Schmid CH, Lau J, Trikalinos TA. Meta-Analyst: software for meta-analysis of binary, continuous and diagnostic data. BMC Med Res Methodol. 2009; 9 (1): 80.

56. Weeks BK, Beck BR. The relationship between physical activity and bone during adolescence differs according to sex and biological maturity. J Osteoporos. 2010; (2010): 1–9.

57. Weeks BK, Young CM, Beck BR. Eight months of regular in-school jumping improves indices of bone strength in adolescent boys and Girls: the POWER PE study. J Bone Miner Res. 2008; 23 (7): 1002–11.

58. Williams MA, Haskell WL, Ades PA, et al. Resistance exercise in individuals with and without cardiovascular disease: 2007 update a scientific statement from the american heart association council on clinical cardiology and council on nutrition, physical activity, and metabolism. Circulation. 2007; 116 (5): 572–84.

59. Yin Z, Gutin B, Johnson MH, et al. An environmental approach to obesity prevention in children: Medical College of Georgia FitKid Project year 1 results. Obes Res. 2005; 13 (12): 2153–61.

60. Yin Z, Moore JB, Johnson MH, et al. The Medical College of Georgia Fitkid project: the relations between program attendance and changes in outcomes in year 1. Int J Obesity. 2005; 29 (S2): S40–5.

61. Zavorsky GS, Longo LD. Exercise guidelines in pregnancy: new perspectives. Sports Med. 2011; 41 (5): 345–60.



© 2014 American College of Sports Medicine


Article Tools



Article Level Metrics

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Connect With Us