The overall findings suggested that exercise may result in clinically relevant benefits to femoral neck and lumbar spine bone mass density in postmenopausal women who had not been participating in exercise levels recommended for bone health. The authors' cautious conclusions reflect the evidence presented and are likely to be reliable.

To assess the effects of ground exercise (such as walking) and/or joint reaction exercise (such as strength training) on femoral neck and lumbar spine bone mineral density in postmenopausal women not participating in exercise levels currently recommended for bone health.

MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), Dissertation Abstracts Online, CINAHL and SPORTDiscus were searched for articles in any language from 1989 to August 2011. Some search terms were reported. Cross-referencing from retrieved studies and previous reviews and handsearches of selected journals (details not reported) were conducted.

Eligible studies were randomised controlled trials (RCTs) of exercise interventions of 24 weeks or more compared to a control group (such as attention control or non-intervention) for postmenopausal women. Eligible women were those who were not regularly participating in any type of regular joint or ground reaction force exercise (definition reported in the paper). Primary outcomes were bone mineral density at the femoral neck and lumbar spines. Secondary outcomes were other measures of bone mineral density, number of fractures, body mass index and a range of other measures (details reported in the review). Included studies had to report bone mineral density data assessed using dual-energy X-ray absorptiometry or dual-photon absorptiometry. Studies of resistance training were included if lower body exercises were part of the programme.

Interventions and comparators varied widely between studies. Studies were conducted in Australia, Canada, Brazil, Japan, Portugal, Sweden, China, Germany, USA and UK. Most of the participants were white or Asian. A number of studies reported concurrent treatment including menopausal hormone therapy, bisphophonates, bisphosphonate or Vitamin D. Several studies included patients with osteoporosis or osteopenia.

Two authors independently selected studies for inclusion. Disagreements were resolved by consensus or recourse to a third reviewer.

Quality was assessed using the Cochrane risk of bias tool for sequence generation, allocation concealment, blinding and reporting of incomplete outcome data. Each item was classed as having high, low or unclear risk of bias.

Two authors independently assessed quality. Disagreements were resolved by consensus.

Data were extracted to enable calculation of Hedge’s standardised effect size (g) adjusted for small sample bias for the difference in change scores between groups. Corresponding 95% confidence intervals were calculated. Data were reported on an intention-to-treat basis.

Two authors independently extracted data. Discrepancies were resolved through consensus or recourse to a third reviewer.

Pooled effect sizes and 95% confidence intervals (CIs) were calculated using a random-effects model. Effect sizes (g) were categorised as trivial (<0.20), small (≥20 to <0.50), medium (≥0.50 to <0.80) or large (≥0.80). The number needed to treat was calculated. Statistical heterogeneity was assessed using the Q and Ι² methods. Ι² values of 50% to <75% were considered medium and 75% or more were considered large. Moderator analyses, meta-regression and sensitivity analyses were used to examine changes in femoral neck and lumbar spine bone mineral density according to more than 20 selected variables. Publication bias was assessed using the Duval and Tweedie trim-and-fill methods.

Twenty-five RCTs (1,775 participants) were included in the review. Most of the studies were considered to be at low risk of bias in terms of sequence generation and blinding and unclear for allocation concealment and incomplete outcome reporting.

Compared to control groups, a small statistically significant improvement was reported for exercise for femoral neck bone mineral density (g 0.288, 95% CI 0.10 to 0.47; Ι²=70.1%; 21 RCTs). There was a difference for lumbar spine bone mineral density but it was not statistically significant (g 0.18, 95% CI -0.003 to 0.361; Ι²=65.3%; 21 RCTs). The number needed to treat was six for femoral neck and lumbar spine bone mineral density outcomes.

No significant effects were reported for the mixed-effects analyses. Results were reported for meta-regression and secondary outcomes.

The overall findings suggested that exercise may result in clinically relevant benefits to femoral neck and lumbar spine bone mass density in postmenopausal women.

The review question was clear with defined inclusion criteria. Several relevant sources were searched and attempts were made to reduce language and publication bias. Appropriate methods were used to reduce reviewer error and bias throughout the review process. An appropriate tool was used to assess quality and results were reported for individual studies. The methods of synthesis appeared appropriate. Significant statistical heterogeneity was found and reasons for this were explored.

The authors' cautious conclusions reflect the evidence presented and are likely to be reliable.

Practice: The authors did not state any implications for practice.

Research: The authors stated that further well-designed randomised controlled trials were needed to evaluate the effects of ground and/or joint reaction force exercise on femoral neck and lumbar spine bone mineral density and fracture risk. Future studies should provide specific details for exercise cut-off points for enrolling participants in studies and should use high resolution peripheral quantitative computed tomography to assess outcomes.

United States Department of Defence, Army Medical Research and Materiel Command.

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.