Critical Factors for Bone Health in Women Across the Age Span: How Important Is Muscle Mass?

Jasminka Ilich-Ernst, RD, MS, PhD, Rhonda A. Brownbill, MS, RD, Martha A. Ludemann, MS, RD, Rongwei Fu, PhD

In This Article


This was a cross-sectional study of healthy white women of wide age range with the aim of assessing various biological and lifestyle factors (age, body composition, Ca intake, and physical activity) possibly influencing bone mass in different skeletal sites. The results showed:

  • Significant reduction of both BMD and LBM with age, as well as the reverse relationship between the age of menarche and number of YSM with BMD of various skeletal sites;

  • Lean body mass was the strongest determinant of BMD in various skeletal sites in the entire population and individual age groups in single and multiple regression models. Its influence in most instances was stronger than that of weight and/or total body fat;

  • Ca intake from food and total Ca (intake from food and supplements) were positively associated with BMD of various regions of hip in the entire cohort and in the youngest and oldest groups (1 and 4). There was no relationship between Ca intake and bone mass in perimenopausal and early postmenopausal women (groups 2 and 3).

  • Measures of lifetime past activity assessed in older women were significantly positively associated with BMD in the total body, spine, hip, and forearm when examined separately and in multiple regression models. There was an added effect of Ca intake acting in concert with physical activity on BMD of Ward's triangle, whereas in other skeletal sites the effect of physical activity was irrespective of Ca intake.

  • Various modes of current walking were positively associated with regions of hip and forearm when examined separately, whereas total hours of present walking were significantly associated with the ultradistal forearm BMD in multiple regression models.

  • Both past activity and present walking modes were associated with a lower body fat and BMI, but there was no relationship with LBM.

The positive relationship between body weight and bone mass in pre- and postmenopausal women has been shown before.[20,21,22] We observed that body weight was positively associated with the BMD of all skeletal sites in our entire population and in some skeletal sites within age groups. However, LBM and, to a lesser extent, TBF exerted stronger influence than weight on BMD of most of the sites in the whole population and in the various groups. Moreover, although the lean and fat tissue was correlated, our subgroup analysis revealed the independent effect of LBM on BMD of various skeletal sites (see Table 3). This effect seems to be similar to that observed in children and adolescents, in whom lean tissue dominates over any other anthropometric parameter.[15,23] These findings imply that LBM as a component of weight may be protective for bones (probably due to the higher load on bones and the positive effect on osteogenesis).

Our data do not support the notion that greater weight or overweight due to the increased TBF would be beneficial for bones unless associated with an accompanying increase in LBM. Increasing weight above the individual's optimal value for height should not be an option for protecting bones and preventing osteoporosis.

Previous studies have shown a significant relationship between muscle strength and bone mass and suggested a relationship between the loss of muscle (sarcopenia) and loss of bone (osteopenia) with age, but have not measured LBM.[5] According to Frost,[24] muscle forces have more influence on bone strength and mass than either genetics or nonmechanical factors (hormones and Ca intake), dominating control of the bone remodeling processes. Muscle exerts about 2 kg of force on bones to move each kg of body weight.

Similarly to bone mass, muscle mass increases during growth, levels off in early adulthood, and gradually declines to less than 60% of that of a young adult by age 75.[25] We were able to show in our population that bone mass decreases as lean body mass decreases and age increases (see Figure 1), and based on the regression equations, a loss of about 0.14 kg of LBM and 13.33 g of BMC occurs for every year of increased age.

The decline in muscle tissue, and consequently strength, in older adults could lower bone strains to the remodeling threshold and result in bone loss.[24] Understanding the relationship between LBM and bone is particularly critical for this population, who almost inevitably experience sarcopenia and related osteopenia. The implementation of measures to engage them in activities that can increase or at least maintain existing LBM is crucial.

In our study, Ca was significantly associated with the trochanter BMD for the entire population when analyzed in multiple regression models. In single regression, Ca was related to femoral neck BMD in the entire population and in group 1 and to the lumbar spine and neck BMD in group 4 (see Figure 3). In addition, as we noted a strong negative association between TBF and Ca intake, we examined their combined influence on BMD of various skeletal sites.

It seems that the possible effect of Ca on BMD is stronger in subjects with lower TBF (see Figure 2). Because the categorization of data leads to substantial reduction in sample size and loss of statistical power, the lack of statistical significance in this case is probably due to inflated type II error. This finding, however, is in agreement with the study by Nguyen and colleagues,[26] who showed similar interaction between Ca and BMI (indicator of obesity) on BMD of femoral neck in a large population of elderly women and men.

Although there is an abundance of literature examining the effects of Ca on bone mass of different skeletal sites in populations of all age groups, it needs to be emphasized that the positive effect of Ca on bones is relatively weakly supported by cross-sectional and observational studies, particularly in younger and early menopausal women, probably as a consequence of strong hormonal effects.[27,28,29] It seems that the effect of dietary Ca on bone is most discernible in the late postmenopausal period.

We showed a positive relationship between Ca and BMD of certain skeletal sites in the youngest and oldest group and in the entire population, but not in perimenopausal and early menopausal women (groups 2 and 3), which, besides the powerful hormonal effect masking any subtle effect of Ca, could also be due to the small group size and large standard deviation in Ca intakes.

On the basis of current evidence, it is clear that the bones of growing individuals benefit the most from various levels of activity and exercise, although it is uncertain whether the benefits will be sustained throughout life. Mixed evidence exists for the effects of vigorous-intensity physical activity on bone mass in postmenopausal women. Evidence for a beneficial effect gained by low- to moderate-intensity activities (eg, walking, heavy housework, gardening) is the least convincing, although these are the most common activities engaged in by older adults. Moreover, the studies examining the possible sustained effects of previous sports and recreational activities on bone mass in older adults are rare, and it is not clear whether such past activity is beneficial to bone mass later in life.

We found past activity (assessed as past engagement in sports and recreational activities) to be significantly correlated with hip, spine, total body, and forearm BMD in our older cohort, whether examined in single or multiple regression models. This may suggest possible sustaining beneficial effects of weight-bearing activities on bones during earlier course of life. Those women who engaged in regular exercise for more than 47% (median percent time for past activity involvement, calculated from age 18 years to present) of their adult life had significantly greater Ward's and forearm BMD. A similar trend (as a result of same stratification) was noted with spine and total body BMD. In addition, when we examined the possible combined influence of Ca and past physical activity, the latter remained dominant, irrespective of a Ca effect on forearm BMD; whereas there was a significant added effect of Ca on BMD of a hip region (see Figure 5).

Our findings are in agreement with the study of Orwoll and coworkers,[30] who found both recent and past activity to be positively associated with lumbar spine and femoral neck BMD in about 8000 women over the age of 65 years. However, Greendale and colleagues[13] failed to observe a significant association between past activity and forearm and spine BMD in postmenopausal women when assessing past activity at specific time points vs assessing total measures of past activity.

It seems that measuring total adult lifetime activity (eg, from 18 years of age) provides a more accurate assessment than measuring activity during specific time periods (eg, in the past 10 years or during young adulthood). It is also possible that women in our cohort who were more active earlier in their lives were also more health-conscious, consumed better self-selected diets, and had healthier lifestyles that all contributed to their better bone mineral status. This inability to tease out the complex interactions occurring throughout life is a major limitation of an observational study.

In our study, walking pace (brisk and fast) and total hours of walking contributed to the variance of hip and forearm BMD, suggesting that walking at a brisk or fast pace benefits bone mass in the hip. Frequency and past walking did not reach statistical significance. Coupland and associates[10] found strong associations between total body and hip BMD and walking and stair climbing. Zystra and colleagues[31] found the number of hours walked per day to be positively related to lumbar spine and femoral neck BMD in women 21-95 years, which is consistent with current views of the importance of mechanical loading on bone.

Although Frost[24] proposes that low-force exercises might be beneficial to bone strength and mass in aging adults (primarily via improving muscle), our data do not support his hypothesis, as we showed only walking at brisk or fast pace benefited bone in hip, implying that the intensity is important. The positive relationship between forearm BMD and total hours of walking (as a measure of lower-intensity exercise) does fit into Frost's hypothesis, as does the notion that even a low-intensity activity might be beneficial for overall strength and fitness.

The limitation of our study is its observational nature and, compared with large epidemiologic studies, a relatively small sample size. Small sample size is particularly a limitation in subgroup analyses in which the number of subjects in each group was even more reduced as a result of stratification. Although observational studies do not allow us to establish a firm cause-and-effect relationship, observing a strong association may make us comfortable in accepting it as such.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.