Authors' objectives

To use a meta-analysis to examine the effects of resistance training on bone mineral density (BMD) at the femur, lumbar spine and radius in pre- and postmenopausal women.

Searching

MEDLINE, Current Contents, SPORTDiscus and Dissertation Abstracts International were searched from January 1966 to December 1998 using the following keywords: 'bone', 'bone density', 'bone mineral density', 'exercise', 'physical activity', 'women', 'females', 'physical fitness', 'fitness', 'weight training', 'resistance exercise', 'resistance training', 'osteoporosis' and 'osteopenia'. The reference lists from both the original and review articles were also reviewed. Selected journals were handsearched, although the names of these journals were not given. Three experts on exercise and BMD reviewed the reference list and coding sheet for thoroughness and completeness.

Study selection

Assessment of study quality

Study quality was assessed using a 3-item questionnaire designed to assess bias, specifically randomisation, blinding and withdrawals or drop-outs. The number of points possible ranged from 0 to 5. All questions were designed to elicit either a yes (1 point) or no (0 point) response. The questionnaire was shown to be both valid (face validity) and reliable: the researcher inter-rater agreement (r) was 0.77 (95% confidence interval: 0.60, 0.86). The authors do not state how the papers were assessed for quality, or how many of the reviewers performed the quality assessment.

Data extraction

All data were extracted independently by two authors. Coding sheets were developed along with instructions that described each code. The authors then met and reviewed every item for accuracy and consistency, and any disagreements were resolved by consensus. The coders were not blinded to study information. The major categories of variables coded included the study characteristics, the physical characteristics of the participants, and the primary and secondary outcomes.

The authors used the standardised difference approach as the ES measure. This was calculated by dividing the difference in the change outcome between the control and exercise groups by the pooled standard deviation of the exercise and control groups. The ES was then corrected for small-sample bias. For studies that included multiple outcomes, the changes in BMD were treated as independent data points. Bootstrap resampling (5,000 iterations) was used to generate 95% bootstrap confidence intervals (BCI) around the mean ES changes for BMD, because of the small sample sizes in this study, especially for subgroup analyses.

The secondary outcomes (changes in body weight, body mass index, percentage of body fat and lean-body mass) were calculated as the difference (exercise minus control) of the changes (initial minus final) in these mean values. The original measure was used for all secondary outcomes.

Methods of synthesis

Results of the review

Twenty-nine studies were included in the review: 18 randomised controlled trials and 11 controlled trials. The total number of participants was 1,097. BMD values were available at the femur (743 participants), at the lumber spine (870 participants), and at the radius (441 participants).

No statistically-significant heterogeneity was found at any of the sites studied, thus, a fixed-effect model was used for the overall results at all three sites.

The mean study quality was 2 (standard deviation, SD: plus or minus 1; range: 1 to 4).

Proximal femur (22 studies).

Small and statistically-insignificant changes in BMD were observed at the femur site. These changes were equivalent to a 0.33% increase in the exercise groups and a 0.05% decrease in the control groups. No evidence of publication bias was observed (r=0.12, P=0.26). The mean ES change in BMD at the femur was 0.07 (SD=0.36; 95% BCI: -0.02, 0.15).

Lumbar spine (23 studies).

Small but statistically-significant ES changes in BMD were found at the lumbar spine. These changes were equivalent to a 0.19% decrease in the exercise groups and a 1.45% decrease in the control groups. No evidence of publication bias was observed (r= -0.08, P=0.62). The mean ES change in BMD at the lumbar spine was 0.24 (SD=0.36; 95% BCI: 0.11, 0.38).

Radius (10 studies).

Small but statistically-significant changes in BMD were observed at the radius. These were equivalent to a 1.22% increase in BMD for the exercise groups and a 0.95% decrease in the control groups. No evidence of publication bias was observed (r=0.17, P=0.38). The mean ES change in BMD at the radius was 0.30 (SD=0.33; 95% BCI: 0.13, 0.48).

With the exception of changes in BMD at the proximal femur, these results were consistent even after each study was deleted from the model once.

Subgroup analysis.

Subgroup analyses found a trend for statistically significantly greater ES changes in BMD at the femur and at the radius for two variables: studies of higher versus lower quality, and postmenopausal versus premenopausal participants. No statistically-significant between-group differences were found for the other variables analysed.

No statistically-significant between-group differences were observed for ES changes at the lumbar spine for any of the variables analysed.

Regression analysis.

Regression analyses found that a change in the percentage of body fat was a significant predictor for ES changes in BMD at the femur. In addition, the initial lean-body mass was a significant predictor for ES changes in BMD at the radius. No significant predictors were observed for ES changes in BMD at the lumbar spine.

Secondary outcomes.

There was a statistically-significant decrease observed in the percentage of body fat (-2% plus or minus 2; 95% BCI: -3, -1), whereas there was a statistically-significant increase in lean-body mass (2 plus or minus 1 kg; 95% BCI: 1, 2). No statistically-significant changes were observed for body weight or body mass index. Muscular strength increased by 40% in the exercise groups and by 6% in the control groups.

Authors' conclusions

The results of this meta-analysis suggested that resistance training has a positive effect on the BMD of all women at the lumbar spine, and in postmenopausal women at the femur and radius.

CRD commentary

The authors stated their review question clearly and the inclusion criteria were well defined. The literature search was well defined and thorough, although foreign language articles were excluded; however, no evidence of publication bias was observed.

Study quality was assessed using a questionnaire, which was shown to be valid and reliable. These findings were used to assess the influence of study quality on the results in a subgroup analysis.

The authors did not report how decisions were made when selecting the studies or conducting the quality assessment.

Details of the studies were tabulated clearly and included most of the important information, such as the study design, the sample size in each group, a description of the interventions, and the outcome assessment measure. However, the participants' characteristics, outcomes, follow-up and withdrawals were not presented for each study. The initial physical characteristics of the participants and the initial BMD values were combined for all trials, and were presented as the mean values and SDs for the exercise and control groups. The BMD results and subgroup analyses, including the 95% BCI and Q statistic, were also presented. The Q statistic was used as a measure of heterogeneity, and was considered significant for values less than 0.05. The authors state that there was no statistically significant heterogeneity at any of the sites observed. These data were supplemented by a narrative discussion.

The authors' conclusions appear justified and the review appears to be relevant to the topic area.

Implications of the review for practice and research

Practice: The authors state that resistance training in conjunction with other types of non-pharmacologic and/or pharmacologic therapy may be most appropriate, especially for those women with osteoporosis.

Research: The authors state that additional studies directed at different lumbar and radius sites would seem appropriate. In addition, the authors suggest that future studies dealing with the effects of resistance training on BMD in women should do a better job of addressing and reporting the dietary habits of their participants, and the types of pharmacologic interventions they may be taking. Furthermore, because few studies included an assessment of the alcohol and calcium intake of the participants, greater attention to these factors seems warranted.

The authors also recommended that future studies include an evaluation of their data using both an analysis-by-protocol and an intention-to-treat approach, because one may consequently examine both the efficacy and effectiveness of resistance training for enhancing BMD in women. They state that this will help provide clinicians with more meaningful information regarding the use of resistance training for enhancing BMD in women. The authors also highlight that additional information on appropriate study designs for examining the effects of exercise on BMD may be found in the review of Snow et al. (see Other Publications of Related Interest no.3). Finally, the authors state that it would seem plausible to suggest that a need exists for a large randomised trial to examine the effect of resistance training on both BMD and fracture risk. They add, however, that a trial of this nature may never be successfully conducted.

Funding

United States Department of Defense, Army Medical Research and Material Command Award 17-98-1-8513.

Bibliographic details

Kelley G A, Kelley K S, Tran Z V. Resistance training and bone mineral density in women: a meta-analysis of controlled trials. American Journal of Physical Medicine and Rehabilitation 2001; 80(1): 65-77. [PubMed: 11138958]

Other publications of related interest

1. Begg CB. Publication bias. In: Copper H, Hedges LV, editors. The handbook of research synthesis. New York: Russell Sage; 1994. p. 399- 409. 2. Hedges LV, Olkin I. Statistical methods for meta-analysis. San Diego (CA): Academic Press; 1985. 3. Snow CM, Matkin CC, Shaw JM. Physical activity and risk for osteoporosis. In: Marcus R, Feldman D, Kelsey J, editors. Osteoporosis. San Diego (CA): Academic Press; 1996. p. 511-28.

Indexing Status

Subject indexing assigned by NLM

MeSH

Aged; Bone Density; Exercise /physiology; Female; Femur /physiology; Humans; Lumbar Vertebrae /physiology; Middle Aged; Postmenopause /physiology; Radius /physiology

AccessionNumber

12001000269

Database entry date

31/07/2002

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.