Journal of Applied Physiology Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 88: 1730-1736, 2000;
8750-7587/00 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (31)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kelley, G. A.
Right arrow Articles by Tran, Z. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kelley, G. A.
Right arrow Articles by Tran, Z. V.
Vol. 88, Issue 5, 1730-1736, May 2000

Exercise and bone mineral density in men: a meta-analysis

George A. Kelley1, Kristi S. Kelley1, and Zung Vu Tran2

1 Department of Kinesiology and Physical Education, Northern Illinois University, DeKalb, Illinois 60115; and 2 Department of Preventive Medicine and Biometrics, University of Colorado Health Sciences Center, Denver, Colorado 80262


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The purpose of this study was to use the meta-analytic approach to examine the effects of exercise on bone mineral density (BMD) in men. A total of 26 effect sizes (ES) representing 225 subjects from 8 studies met the criteria for inclusion. When BMD sites assessed were specific to the sites loaded during exercise, increases of ~2.6% (2.1% in the exercisers and -0.5% in the controls) were found. These results were statistically significant (ES = 0.213, 95% bootstrap confidence interval = 0.007-0.452). Statistically significant ES changes were found for older (>31 yr) but not younger (<31 yr) adults, with differences between groups statistically significant (P = 0.04). Statistically significant changes were also observed at the femur, lumbar, and os calcis sites. The results of this study suggest that site-specific exercise may help improve and maintain BMD at the femur, lumbar, and os calcis sites in older men. However, the biological importance of the small changes observed for most outcomes, quality of studies, and limited data pool prevent us from forming any firm conclusion regarding the use of exercise for maintaining and/or improving BMD in men. Clearly, a need exists for additional studies.

men; osteoporosis; systematic review; review; physical activity


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

OSTEOPOROSIS AND LOW BONE mass are major public health problems affecting ~23,000,000 women >= 50 yr of age in the United States (28). However, osteoporosis and low bone mass are also a major public health problem among men, affecting ~5,000,000 individuals >= 50 yr of age in the United States (28). Approximately 5% of white, Asian, Hispanic, and American-Indian men 50-79 yr of age have osteoporosis, whereas 3.5% of black men 50-79 yr of age have the disease (28). For those individuals >= 80 yr of age, these numbers increase to ~24% among white men, 17% among black men, and 5% among Asian, Hispanic, and American-Indian men (28). In addition, it has been estimated that osteoporosis-related fractures represent 3% of all Medicare costs and that the lifetime risk of an osteoporotic fracture for white men >= 50 yr of age is ~13% (33).

However, this lifetime risk may be an underestimate, inasmuch as a recent study in Australia found that the residual lifetime fracture risk in 60-yr-old men with average life expectancy was 29% (18).

Exercise has been recommended as a nonpharmacological approach for maximizing bone mineral density (BMD) during the younger years as well as improving bone density by increasing and/or preventing the loss of bone during the older years (35). Exercise may be especially appropriate, since it is a low-cost intervention that is available to most of the general public. However, training studies examining the effects of exercise on BMD in men have led to conflicting results (2, 4-6, 13, 23-25, 31, 36). For example, of the 10 studies previously cited, only 39% of the sites assessed were reported as statistically significant and positive compared with a control group. One of the possible reasons for the lack of statistically significant results may be the small sample size that comprised many of these trials. As a result, the ability to detect meaningful differences may have been compromised. Meta-analysis is a quantitative approach in which individual study findings addressing a common problem are statistically integrated and analyzed (8, 14, 16, 29). It has been shown to be more accurate than the vote-counting approach (15) and may be especially useful when the number of studies is small and/or the sample sizes within each study are small (29). Although several meta-analyses have been conducted on the effects of exercise on bone density in women (3, 19-21, 37), we are not aware of any meta-analytic work dealing with the effects of exercise on BMD in men. Given the health-care consequences of low BMD in men and the potential for exercise to improve BMD, a need exists to use a quantitative approach to examine the effects of exercise on BMD in this population. Thus the purpose of this study was to use the meta-analytic approach to examine the effects of exercise on BMD in men.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Data Sources

Computerized literature searches of articles indexed between January 1966 and December 1998 were performed using MEDLINE, Current Contents, Sport Discus, and Dissertation Abstracts International databases. The following keywords were used alone or in various combinations for computer searches: bone, exercise, physical activity, men, males, physical fitness, fitness, and osteoporosis. The titles and abstracts of studies identified in the computerized searches were examined to exclude any that were clearly irrelevant. The full text of the remaining articles was retrieved, and each paper was read to determine whether it contained information on the topic of interest. Because computer searches have been shown to yield fewer than two-thirds of relevant articles (9), the reference lists from original and review articles were also reviewed to identify any studies that had not been previously identified and appeared to contain information on the topic of interest. Hand searching of selected journals was also performed. Furthermore, three experts on exercise and BMD (Dr. Charlotte Sanborn, Dr. David Nichols, and Dr. Christine Snow) reviewed our reference list and coding sheet for thoroughness and completeness.

Study Selection

Inclusion criteria for this study were as follows: 1) randomized or nonrandomized trials that included a comparative nonexercise control group, 2) exercise as the only intervention, 3) adult men (mean study age >= 18 yr) as subjects, 4) journal articles, dissertations, and masters theses published in the English-language literature, 5) studies published and indexed between January 1966 and December 1998, 6) BMD (relative value of bone mineral per measured bone area) assessed, and 7) training studies lasting >= 16 wk. Only information that met the above criteria was included in our analysis. Thus, for example, if BMD was also assessed in women, we did not include this information, since it did not meet our inclusion criteria. Because dissertations may eventually become full-length journal articles, we cross-referenced between the two to avoid duplication. We did not include abstracts and conference papers from national meetings because of the paucity of data provided as well as the inability to obtain complete data from the authors. Studies published in foreign language journals were also not included because of the potential error in the translation and interpretation of findings. Studies that met our inclusion criteria were also examined to ensure that the same subjects were not included in more than one study (11). For the two studies that met our inclusion criteria but did not provide appropriate information on changes in BMD (4, 23), personal contact was made with the authors to retrieve such information.

Data Extraction

Coding sheets that could hold 242 items were developed and utilized in this investigation. In addition, coding instructions that described how to code each item on the coding sheet were developed and utilized. To avoid inter- and intracoder bias, all data were independently extracted by two authors. The authors then met and reviewed every data point for accuracy and consistency. Disagreements were resolved by consensus. The major categories of variables coded included study characteristics, physical characteristics of subjects, and primary and secondary outcomes.

Statistical Analysis

Primary outcomes. The primary outcome in this study was changes in BMD. Because of the various ways in which the authors reported data on changes in BMD and because we also wanted to maximize the number of studies and outcomes that could be included in our analysis, we used the standardized difference approach as our effect size (ES) measure. This measure provides one with a statistic similar to a z-score. Each ES was calculated by subtracting the change outcome (percent or absolute) in the exercise group from the change outcome in the control group and then dividing this difference by the pooled standard deviation of the exercise and control groups (16). The ES was then corrected for small-sample bias (16). For those studies that did not report change outcome variances, these were estimated using previously developed methods (12). In general, an ES of 0.20 is considered a small effect, 0.50 a moderate effect, and 0.80 a large effect (7). An ES of 0.50, for example, means that the exercise group differed from the control group by 0.5 SD in favor of the exercise group. By use of a z-score table, this means that the exercise group would do better than ~69% of the control group. Although the first metric of choice should be the one that has the most meaning to the reader, in this case, the percent change difference, we were unable to use this metric because of insufficient reporting of data for calculating within-group percent change variances for each study. However, to enhance interpretation, we also calculated percent change differences for each study. Because of the small sample size in this study, especially for subgroup analyses, bootstrap resampling (5,000 iterations) was used to generate 95% bootstrap confidence intervals (BCI) around mean ES changes for BMD (10). The bootstrap technique is a computer-intensive, nonparametric method of estimating the reliability of the original sample estimate, in this case, ES changes in BMD. By randomly drawing from the available sample, with replacement, samples the same size as the original are generated. Each time an observation is selected for a new sample, each of the elements of the original sample has an equal chance of being selected. This is similar to replicating each member of a sample 5,000 times (iterations). The main advantage of this approach is that the estimate desired is not based on some theoretical distribution but, rather, on the sample itself. This approach frees one from the constraints of the central limit theorem. The number of iterations chosen was based on previous research demonstrating that improvement of estimation accuracy was limited beyond 5,000 iterations (38). If the 95% confidence interval included zero (0.00), it was concluded that there was no statistically significant effect of exercise on BMD.

Heterogeneity of ES changes in BMD was examined using the Q statistic (16). A random-effects model was used if changes were significantly heterogeneous (P <=  0.05), whereas a fixed-effects model was used in the absence of significant heterogeneity (26).

For studies that included multiple outcomes because of more than one group, net changes were initially treated as independent data points. However, to examine the influence (sensitivity) of each study on the overall results, analyses were performed with each study deleted from the model.

Publication bias (the tendency for journals and/or authors to publish studies that yield statistically significant results) was examined using Kendall's tau  statistic (1). A statistically significant result (P <=  0.05) was considered to be suggestive of publication bias. In addition, we used a semiquantitative approach (funnel plot) to examine potential publication bias (22). This was accomplished by plotting the sample size on the vertical axis and changes in blood pressure on the horizontal axis. Usually, smaller studies will be more dispersed at the bottom of the funnel, whereas larger studies will be more congregated at the top. A gap at the bottom of the funnel on the left side indicates that small studies yielding null or negative results may be missing.

Study quality was assessed using a three-item questionnaire designed to assess bias, specifically, randomization, blinding, and withdrawals/dropouts (17). The number of points possible ranged from a low of 0 to a high of 5. All questions were designed to elicit yes (1 point) or no (0 points) responses. The questionnaire took <10 min/study. The questionnaire has been shown to be valid (face validity) and reliable (researcher-interrater agreement, r = 0.77, 95% confidence interval = 0.60-0.86) (17).

Subgroup analyses. Subgroup analyses for primary outcomes were performed using ANOVA-like procedures for meta-analysis (16). These procedures provide statistics for within (Qw)- and between (Qb)-group differences. If statistically significant within-group (Qw) heterogeneity existed (P < 0.05), a random-effects model was used. If no statistically significant within-group (Qw) heterogeneity existed, a fixed-effects model was used. ES changes in BMD were initially examined when the data were partitioned according to whether the BMD sites assessed were specific to the sites loaded during the exercise protocol. Subgroup analysis was also performed when the data were partitioned according to the age of the subjects (<= 31 vs. >31 yr of age), limited to those results in which the BMD sites assessed were specific to the sites loaded. We chose 31 yr of age as our cut point because it has been reported that peak bone mass is attained as late as ~30 yr of age (32). Further analyses were performed on site-specific results for older subjects in relation to specific assessment location (femur, lumbar, os calcis) and study design (randomized controlled trial vs. nonrandomized controlled trial). Bootstrap resampling (5,000 iterations) (10) was used to generate 95% confidence intervals around ES changes for all subgroups. Randomization tests (5,000 iterations) were used to generate probability values for between-group differences (30). Randomization tests using 5,000 iterations can detect a probability as low as 0.002 (30).

Regression analysis. Potential associations between ES changes in BMD and initial BMD and length of training were examined using regression procedures previously described by Hedges and Olkin (16).

Secondary outcomes. Secondary outcomes (changes in body weight, body mass index, percent body fat, lean body mass, maximum oxygen consumption, resting heart rate, calcium intake) were calculated as the difference (exercise minus control) of the changes (initial minus final) in these mean values. The original metric was used for all secondary outcomes. For those studies in which variance estimation was necessary, these were accomplished using the same procedures used to estimate variances for BMD (12). Fixed- and random-effects models were used following the same procedures described for BMD.

Unless otherwise noted, values are means ± SD. The alpha -level for statistical significance was set at P <=  0.05. Bonferroni adjustments were not made because of the increased risk of a type II error.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Study Characteristics

A total of 3,141 titles and abstracts were reviewed. From those, 26 ES representing a total of 225 subjects (135 exercise, 90 control) and 18 groups (10 exercise, 8 control) from 8 studies met the criteria for inclusion (2, 5, 6, 13, 24, 25, 31, 36). The per person time to code each study once ranged from 0.67 to 2.63 h (1.04 ± 0.66 h). Two studies were not included because we were unable to obtain data necessary for the calculation of an ES (4, 23). Thus our percent loss that met our inclusion criteria was 20. One study (34) was not included because it contained some of the same subjects from another study that met our inclusion criteria (24). We chose to include the latter study because more complete data were available for extraction. A general description of the studies is shown in Table 1. Four studies were conducted in the United States (5, 24, 25, 36), two in Australia (2, 31), and one each in Japan (13) and the United Kingdom (6). For the five studies that reported such information (2, 6, 24, 31, 36), percent dropout, defined as the number of subjects who did not complete the study, ranged from 0 to 31% in the exercise groups (10 ± 12%) and from 0 to 22% in the control groups (5 ± 10%). The number of subjects in which pre- and post-BMD measures were assessed and included in our analysis ranged from 3 to 28 in the exercise groups (14 ± 8) and from 7 to 21 in the control groups (11 ± 6). Study quality ranged from 0 to 3 (1 ± 1). Compliance, defined as the percentage of exercise sessions attended, was >90% for the two studies that reported this information (13, 24). Reliability for BMD assessment (coefficient of variation) ranged from 0.4 to ~5.0% (1.5 ± 1.0%).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   General description of included studies

Physical Characteristics of Subjects

A description of the physical characteristics for the exercise and control groups may be found in Table 2. Only one study reported that calcium supplementation was given to subjects (13). For the four studies that reported such information, three reported that none of the subjects was taking any type of pharmacological intervention other than calcium that could affect BMD (2, 6, 13), and one (5) reported that subjects were taking drugs that could affect BMD. For the two studies that reported information on cigarette smoking, one reported that none of the subjects smoked cigarettes (24), and another reported that some of the subjects smoked (13). Three studies reported that subjects had not been physically active before participation in the study (5, 13, 24), whereas it appeared that another study included subjects that had been previously active (2). The two studies that reported information on previous fractures reported that subjects had not had previous fractures at the site(s) assessed before participation in the study (24, 25). None of the studies reported information on the alcohol intake of the subjects.

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Initial physical characteristics

Primary Outcomes

Individual ES changes for primary outcomes (changes in BMD) are shown in Table 3. Approximately 39% of the 26 ES were reported as statistically significant and positive by the authors. Initial BMD values for all sites assessed ranged from 0.700 to 1.400 g/cm2 in the exercise groups (1.120 ± 0.188 g/cm2) and from 0.670 to 1.290 g/cm2 in the control groups (1.078 ± 0.168 g/cm2). With use of a random-effects model because of statistically significant heterogeneity (Q = 45.01, P = 0.008), overall ES changes were not statistically significant (ES = 0.028, 95% BCI = -0.166 to 0.230). ES changes were equivalent to an exercise-minus-control improvement of 2% (1.6% increase in the exercising subjects and 0.4% decrease in the controls). With each study deleted from the model once, ES changes ranged from a low of 0.028 (95% BCI = -0.166 to 0.230) to a high of 0.199 (95% BCI = -0.131 to 0.489). Although there was no quantitative evidence supporting publication bias (r = -0.23, P = 0.10), funnel plot analysis was suggestive of this potential form of bias.

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Individual ES from studies

Subgroup Analyses

When data were partitioned according to whether the BMD sites assessed were specific to the sites loaded during the exercise protocol, statistically significant within-group changes were found when the sites assessed were specific to the site loaded (ES = 0.213, 95% BCI = 0.007-0.452) but not when the sites assessed were not specific to the sites loaded (ES = -0.205, 95% BCI = -0.613 to 0.219). ES changes in BMD were equivalent to exercise-minus-control increases of ~2.6% (2.1% in the exercisers and -0.5% in the controls) when the sites assessed were specific to the sites loaded and ~0.3% (-0.06% in the exercisers and -0.3% in the controls) when the sites assessed were not specific to the sites loaded. Although it was not statistically significant, there was a trend for between-group changes to be greater when the sites assessed were specific to the sites loaded (Qb = 3.02, P = 0.10). Further subgroup analyses were then performed by limiting analyses to only those sites in which the assessment of BMD was specific to the sites loaded. These are shown in Table 4. Within-group analysis demonstrated that statistically significant increases were found for older but not younger adults and that the differences between groups were statistically significant. These changes were equivalent to an exercise-minus-control improvement of ~6.7% for older subjects (4.2% increase in exercise groups and 2.5% decrease in controls) and 0.4% increase in younger subjects (1% increase in exercisers and 0.6% increase in controls). When the one study that yielded two ES from heart transplant patients in the older group was deleted from the model, statistically significant within-group effects were again observed for older (ES = 0.442, 95% BCI = 0.207-0.799) but not younger (ES = 0.066, 95% BCI = -0.158 to 0.333) subjects. However, no statistically significant differences were observed between groups (Qb = 2.26, P = 0.23). ES changes were equivalent to exercise-minus-control improvements of ~4.0% in older subjects (2% increase in exercisers and 2% decrease in controls). With ES results limited to only older adults, statistically significant within-group ES were found at the femur, lumbar, and os calcis sites, with no statistically significant between-group differences observed. ES changes in BMD were equivalent to exercise-minus-control improvements of ~5.9% at the femur site (4% increase in exercisers and 1.9% decrease in controls), 10.7% at the lumbar site (5.8% increase in exercisers and 4.9% decrease in controls), and 1.6% at the os calcis site (2.1% increase in exercisers and 0.5% increase in controls). With the one study in heart transplant patients deleted, there was an exercise-minus-control improvement of ~6.1% in lumbar BMD (1.0% in exercisers and -5.1% controls). Statistically significant within-group ES increases in BMD were also found when data were partitioned by study design. This was equivalent to exercise-minus-control improvements of ~13.5% for randomized trials (9.8% increase in exercisers and 3.7% decrease in controls) and 4.2% for nonrandomized trials (2% increase in exercisers and 2.2% decrease in controls). However, the percent changes found for randomized controlled trials were derived from the one study in heart transplant patients. No statistically significant ES differences were observed between groups.

                              
View this table:
[in this window]
[in a new window]
  		Table 4.   Subgroup analyses

Regression Analysis

No statistically significant associations were found between ES changes in BMD and length of training or initial BMD.

Secondary Outcomes

No statistically significant differences were found for any of the secondary outcomes. These included body weight (-0.3 kg, 95% BCI = -2.2 to 0.79 kg), body mass index (-0.9 kg/m2, 95% BCI = -0.9 to 0.1 kg/m2), and lean body mass (0.3 kg, 95% BCI = -0.2 to 0.6). Insufficient data were provided to assess changes in percent fat, maximum oxygen consumption, resting heart rate, and calcium intake.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The results of this study suggest that site-specific exercise may help improve and maintain BMD in older men. These results also support the notion that changes in BMD are specific to the sites loaded during exercise in older men. However, these results should be interpreted with caution, inasmuch as they were based on a very limited data pool. Perhaps little change was seen in BMD among younger subjects because subjects already possessed optimal levels of BMD and/or the loss of BMD generally occurs during the later, rather than during the younger, years. In addition, it may also be that more subjects in the older group had been sedentary for a longer period or were not able to ambulate very much. This may be especially true for the one study that included heart transplant patients, because when we deleted this study from our subgroup analysis the increase in BMD in the exercise group decreased from ~6% to 1% (5).

The fact that exercise is an inexpensive, nonpharmacological approach that is available to most of the general public makes this form of treatment appealing, especially given the other physiological and psychological benefits that may be derived from participation in exercise. However, it is important to realize that ~60% of adults in the United States do not regularly participate in adequate amounts of physical activity (27). Furthermore, only 16% of the US population between the ages of 18 and 64 yr report that they regularly participate in progressive-resistance exercise (27). In addition, one must also consider the potential adverse effects of exercise, e.g., arthritis, injury, and cardiac events. Unfortunately, we were unable to determine any potential adverse effects of exercise in this meta-analysis because only one study, limited to heart transplant patients, reported such information (5). The authors concluded that the exercise intervention was safe because it was not associated with an increase in rejection (5). Thus, although participation in a regular exercise program may be efficacious for improving and/or maintaining BMD in men, it may not be very effective in the "real world." Consequently, alternative nonpharmacological and pharmacological interventions for increasing and/or maintaining BMD may be necessary.

Despite the positive results observed in this study, the biological importance of these small changes might not be sufficient to recommend exercise alone as a nonpharmacological intervention. Because the primary reason for improving and/or maintaining BMD is to reduce fracture risk, one would like to know how much of an increase or prevention of loss in BMD is necessary to reduce the incidence of fracture. A recent longitudinal study has shown that femoral neck BMD in men was 24% lower in those with hip fractures and 12% lower in those with fractures of the vertebrae and upper limb (28a). The study also found that a decrease of 1 SD in femoral neck BMD was associated with a 2.3, 1.9, and 1.5 increase in the odds of fracture at the hip, vertebrae, and upper limbs, respectively (28a). The smaller results observed in our study suggest that if exercise is recommended, it should be done only in conjunction with other types of nonpharamcological and/or pharmacological interventions. Furthermore, we are not aware of any consensus as to how exercise-induced increases in BMD affect bone strength. For example, the implications of bone mass laid down on periosteum vs. endosteum may differ in relation to altering bone strength.

A second factor that warrants caution in the interpretation of our results is the relatively low quality of the studies included in our analysis. For example, only one study received a score of 3 while the remaining studies received scores ranging from 0 to 2. In addition, the different types of exercise interventions varied considerably. Furthermore, the fact that our funnel plot analysis was indicative of publication bias suggests that the results of this study may be an overestimate of the effects of exercise on BMD in men. Ideally, to examine the efficacy and effectiveness of exercise on BMD in men, it is suggested that future studies 1) randomize subjects to an exercise and control condition, 2) blind the person responsible for the assessment of BMD to the treatment assignment of subjects, 3) limit participation to only subjects who have been previously sedentary, 4) include weight-bearing exercise protocols that are reflective of those in which the population will be able to participate, 5) assess BMD at the sites that were loaded during the exercise protocol, and 6) include efficacy as well as effectiveness (intention-to-treat) analysis. In addition, it would be beneficial to examine any changes in BMD that might occur on cessation of exercise. Furthermore, it is critical that authors submit and be allowed to publish well-designed studies that yield null results. Consequently, we will be able to form a more valid conclusion regarding the true effects of exercise on BMD in men. Given the prevalence of low BMD in men >= 50 yr of age, it may be especially important to focus on this population (28).

Another factor that warrants caution in the interpretation of our results was the availability of data. Although meta-analysis is more quantitative than traditional narrative reviews, potential problems exist. The meta-analytic review, like any review, is limited by the available data. One potential problem in this investigation was the small number of ES available for some of the subgroup analyses. For example, ES changes in BMD at the specific sites we were able to examine (femur, lumbar, os calcis) were limited to two or three outcomes per site.

Despite our resampling approach because of the small sample sizes available for many of the analyses, additional studies in this area are needed before any firm conclusions can be made regarding the efficacy and effectiveness of exercise as a nonpharmacological intervention for improving and/or maintaining BMD in men. In addition, insufficient information was available to examine results according to the impact of the exercise protocols on the BMD sites assessed. Future studies need to provide a complete description of the exercise intervention.

In conclusion, the results of this study suggest that site-specific exercise may help increase and/or maintain BMD at the femur, lumbar, and os calcis sites in older men. However, the biological importance of the relatively small changes observed for most outcomes, quality of studies included, and limited data pool prevent us from forming any firm conclusion regarding the use of exercise for maintaining and/or improving BMD in men.


    ACKNOWLEDGEMENTS

This study was funded by US Department of Defense, Army Medical Research and Material Command Award 17-98-1-8513.


    FOOTNOTES

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests and other correspondence: G. A. Kelley, Meta-Analytic Research Group, Dept. of Kinesiology and Physical Education, Anderson Hall, Rm. 233, Northern Illinois University, DeKalb, IL 60115-2854 (E-mail: gkelley{at}niu.edu).

Received 27 July 1999; accepted in final form 10 December 1999.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Begg, CB. Publication bias. In: The Handbook of Research Synthesis, edited by Cooper H, and Hedges LV.. New York: Russell Sage, 1994, p. 399-409.

2.   Bennell, KL, Malcolm SA, Khan KM, Thomas SA, Reid RJ, Brukner PD, Ebeling PR, and Wark JD. Bone mass and bone turnover in power athletes and controls: a 12-month longitudinal study. Bone 20: 477-484, 1997[Medline].

3.   Berard, A, Bravo G, and Gauthier P. Meta-analysis of the effectiveness of physical activity for the prevention of bone loss in postmenopausal women. Osteoporos Int 7: 331-337, 1997[Web of Science][Medline].

4.   Blumenthal, JA, Emery CF, Madden DJ, Schniebolk S, Riddle MW, Cobb FR, Higgenbotham M, and Coleman RE. Effects of exercise training on bone density in older men and women. J Am Geriatr Soc 39: 1065-1070, 1991[Web of Science][Medline].

5.   Braith, RW, Mills RM, Welsch MA, Keller JW, and Pollock ML. Resistance training restores bone mineral density in heart transplant patients. J Am Coll Cardiol 28: 1471-1477, 1996[Abstract].

6.   Cohen, B, Millet PJ, Mist B, Laskey MA, and Rushton N. Effect of exercise training programme on bone mineral density in novice college rowers. Br J Sports Med 29: 85-88, 1995[Abstract/Free Full Text].

7.   Cohen, J. Statistical Power Analysis for the Behavioral Sciences. New York: Academic, 1988, p. 1-17.

8.   Cooper, HC, and Hedges LV (Editors). The Handbook of Research Synthesis. New York: Russell Sage, 1994, p. 3-14.

9.   Dickersin, K, Hewett P, Mutch L, Chalmers I, and Chalmers TC. Perusing the literature: comparison of Medline searching with a perinatal trials database. Control Clin Trials 6: 306-317, 1985[Web of Science][Medline].

10.   Efron, B, and Tibshirani R. An Introduction to the Bootstrap. London: Chapman & Hall, 1993, p. 10-15.

11.   Egger, M, and Davey Smith G. Meta-analysis: bias in location and selection of studies. Br Med J 316: 61-66, 1998[Free Full Text].

12.   Follmann, D, Elliot P, Suh I, and Cutler J. Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol 45: 769-773, 1992[Web of Science][Medline].

13.   Fujimura, R, Ashizawa N, Watanabe M, Mukai N, Amagai H, Fukubayashi T, Hayashi T, Tokuyama K, and Suzuki M. Effect of resistance exercise training on bone formation and resorption in young male subjects assessed by biomarkers of bone metabolism. J Bone Miner Res 12: 656-662, 1997[Web of Science][Medline].

14.   Glass, GV, McGaw B, and Smith ML. Meta-Analysis in Social Research. Newbury Park, CA: Sage, 1981, p. 21-24.

15.   Hedges, LV, and Olkin I. Vote-counting methods in research synthesis. Psychol Bull 88: 359-369, 1980[Web of Science].

16.   Hedges, LV, and Olkin I. Statistical Methods for Meta-Analysis. San Diego, CA: Academic, 1985, p. 13, 78, 122-130, 149-165, 167-183.

17.   Jadad, AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, and McQuay HJ. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 17: 1-12, 1996[Web of Science][Medline].

18.   Jones, G, Nguyen T, Sambrook PN, Kelly PJ, Gilbert C, and Eisman JA. Symptomatic fracture incidence in elderly men and women: the Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporos Int 4: 277-282, 1994[Web of Science][Medline].

19.   Kelley, GA. Aerobic exercise and bone density at the hip in postmenopausal women: a meta-analysis. Prev Med 27: 798-807, 1998[Web of Science][Medline].

20.   Kelley, GA. Aerobic exercise and lumbar spine bone mineral density in postmenopausal women: a meta-analysis. J Am Geriatr Soc 46: 143-152, 1998[Web of Science][Medline].

21.   Kelley, GA. Exercise and regional bone mineral density in postmenopausal women: a meta-analytic review of randomized trials. Am J Phys Med Rehabil 77: 76-87, 1998[Web of Science][Medline].

22.   Light, RJ, and Pillemer DB. Summing Up: The Science of Reviewing Research. Cambridge, MA: Harvard University Press, 1984, p. 65-69.

23.   McCartney, N, Hicks AL, Martin J, and Webber CE. Long-term resistance training in the elderly: effects on dynamic strength, exercise capacity, muscle, and bone. J Gerontol A Biol Sci Med Sci 50: B97-B104, 1995[Abstract].

24.   Menkes, A, Mazel S, Redmond RA, Koffler K, Libanati CR, Gundberg CM, Zizic TM, Hagberg JM, Pratley RE, and Hurley BF. Strength training increases regional bone mineral density and bone remodeling in middle-aged and older men. J Appl Physiol 74: 2478-2484, 1993[Abstract/Free Full Text].

25.   Michel, BA, Lane NE, Bjorkengren A, Bloch DA, and Fries JF. Impact of running on lumbar bone density: a 5-year longitudinal study. J Rheumatol 19: 1759-1763, 1992[Web of Science][Medline].

26.   Mosteller, F, and Colditz GA. Understanding research synthesis (meta-analysis). Annu Rev Public Health 17: 1-23, 1996[Web of Science][Medline].

27.   National Center for Health Statistics. Healthy People 2000 Review, 1997. Hyattsville, MD: US Public Health Service, 1997.

28.   National Osteoporosis Foundation. 1996 and 2015 Osteoporosis Prevalence Figures: State by State Report. Washington, DC: Natl. Osteoporos. Found., 1997.

28a.   Nguyen, TV, Eisman JA, Kelley PJ, and Sambrook PN Risk factors for osteoporotic fractures in elderly men. Am J Epidemiol 144: 255-263, 1996[Abstract/Free Full Text].

29.   Petitti, DB. Meta-Analysis, Decision Analysis, and Cost-Effectiveness Analysis: Methods for Quantitative Synthesis in Medicine. New York: Oxford University Press, 1994, p. 15-20.

30.   Potvin, C, and Roff D. Distribution-free and robust statistical methods: viable alternatives to parametric statistics? Ecology 74: 1617-1628, 1993[Web of Science].

31.   Pritchard, JE, Nowson CA, and Wark JD. Bone loss accompanying diet-induced or exercise-induced weight loss: a randomised controlled study. Int J Obes 20: 513-520, 1996.

32.   Recker, RR, Davies M, Hinders SM, Heaney RP, Stegman MR, and Kimmel DB. Bone gain in young adult women. JAMA 268: 2403-2408, 1992[Abstract/Free Full Text].

33.   Riggs, BL, and Melton LJ. The worldwide problem of osteoporosis: insights afforded by epidemiology. Bone 17, Suppl: 505S-511S, 1995[Medline].

34.   Ryan, AS, Treuth MS, Rubin MA, Miller JP, Nicklas BJ, Landis DM, Pratley RE, Libanati CR, Gundberg CM, and Hurley BF. Effects of strength training on bone mineral density: hormonal and bone turnover relationships. J Appl Physiol 77: 1678-1684, 1994[Abstract/Free Full Text].

35.   Snow-Harter, C, and Marcus R. Exercise, bone mineral density, and osteoporosis. Exerc Sport Sci Rev 10: 351-388, 1991.

36.   Williams, JA, Wagner J, Wasnich R, and Heilbrun L. The effect of long-distance running upon appendicular bone mineral content. Med Sci Sports Exerc 16: 223-227, 1984[Web of Science][Medline].

37.   Wolff, I, van Croonenborg JJ, Kemper HCG, Kostense PJ, and Twisk JWR The effect of exercise training programs on bone mass: a meta-analysis of published controlled trials in pre- and postmenopause women. Osteoporos Int 9: 1-12, 1999[Web of Science][Medline].

38.   Zhu, W. Making bootstrap statistical inferences: a tutorial. Res Q Exerc Sport 68: 44-55, 1997[Web of Science][Medline].

J APPL PHYSIOL 88(5):1730-1736
8570-7587/00 $5.00 Copyright © 2000 the American Physiological Society


This article has been cited by other articles:


Home page
 Sports Health: A Multidisciplinary ApproachHome page
D. Leigey, J. Irrgang, K. Francis, P. Cohen, and V. Wright
Participation in High-Impact Sports Predicts Bone Mineral Density in Senior Olympic Athletes
Sports Health: A Multidisciplinary Approach, November 1, 2009; 1(6): 508 - 513.
[Abstract] [Full Text] [PDF]

Home page
 Sports Health: A Multidisciplinary ApproachHome page
S. L. Manske, C. R. Lorincz, and R. F. Zernicke
Bone Health: Part 2, Physical Activity
Sports Health: A Multidisciplinary Approach, July 1, 2009; 1(4): 341 - 346.
[Abstract] [Full Text] [PDF]

Home page
 AMERICAN JOURNAL OF LIFESTYLE MEDICINEHome page
S. B. Going and M. Laudermilk
Osteoporosis and Strength Training
American Journal of Lifestyle Medicine, July 1, 2009; 3(4): 310 - 319.
[Abstract] [PDF]

Home page
 Arch Intern MedHome page
W. Kemmler, D. Lauber, J. Weineck, J. Hensen, W. Kalender, and K. Engelke
Benefits of 2 Years of Intense Exercise on Bone Density, Physical Fitness, and Blood Lipids in Early Postmenopausal Osteopenic Women: Results of the Erlangen Fitness Osteoporosis Prevention Study (EFOPS)
Arch Intern Med, May 24, 2004; 164(10): 1084 - 1091.
[Abstract] [Full Text] [PDF]

Home page
 Endocr. Rev.Home page
M. Peacock, C. H. Turner, M. J. Econs, and T. Foroud
Genetics of Osteoporosis
Endocr. Rev., June 1, 2002; 23(3): 303 - 326.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (31)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kelley, G. A.
Right arrow Articles by Tran, Z. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kelley, G. A.
Right arrow Articles by Tran, Z. V.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online