Physical Activity During Life Course and Bone Mass

A Systematic Review of Methods and Findings From Cohort Studies With Young Adults

Renata M Bielemann; Jeovany Martinez-Mesa; Denise P Gigante


BMC Musculoskelet Disord. 2013;14(77) 

In This Article


Nineteen manuscripts met inclusion criteria. Lumbar spine was the skeletal site most studied (n = 15). Different questionnaires were used for physical activity evaluation. Peak strain score was also used to evaluate physical activity in 5 manuscripts. Lack of statistical power calculation was the main problem found in the quality assessment of all studies. More positive associations between physical activity and bone mass were found in males than in females and when physical activity measurements were done from adolescence to adulthood – than when evaluated in only one period.

This is the first study to systematically review the literature about cohort studies that evaluated the effect of physical activity on bone mass measurements in young adults. The choice of this age group was based on the scarcity in the literature about this subject in individuals who are in the maintenance phase of bone mass, since several studies performed in adults are in pre or menopausal women and older people, who have an increased risk for hip fracture. The main strength of this study was the selection of articles performed independently by two reviewers and a third revision in case of disagreement. This strategy reduces the possibility that some important article might not be identified. Another positive aspect of this study was the quality assessment of these papers which helps to detect fragilities of each included study.

We found 11 different samples with prospective physical activity evaluation and bone mass measurements in young adulthood. It was difficult to summarize findings from the 19 included studies, since there is much heterogeneity among them. The sources of heterogeneity were anatomical sites of bone mass measurements, the evaluated genders and mainly different physical activity assessments.

The quality assessment by Downs & Black criterion[10] showed that the most important aspect found was the lack of statistical power analysis, since no manuscripts reported the power calculation. In addition, as the most part of the studies had a sample size lower than 200 subjects, it is possible that some analyses were not statistically significant due to their low statistical power. Another important aspect in the quality assessment was that the characteristics of the losses were not described in some manuscripts. On the other hand, all studies included at this review used in the analysis adjustment at least for the body size, evidencing the authors' concern with the statistical analyses.

Findings from the studies included showed that around half of the analyses using lumbar spine or femoral neck bone mass as outcome were positively significant, whereas only one third of them were positively significant for total body bone mass. One explanation for this fact is that bone adaptation is limited to loaded regions.[5] Other reason could be the fact that the total body site also includes no weight bearing anatomical sites, such as the wrist, and the majority of physical activities practiced by healthy individuals are weight bearing (walking, running, etc.) and specific activities such as handball and weight lifting are less practiced.[47] These arguments could explain why higher percentages of these analyses were positively associated with weight bearing sites (lumbar spine and femoral neck) than with total body.

The bone mass peak is prior to age 20 years at the proximal femoral sites and 6 to 10 years later for total skeletal mass.[3] So, it would be expected that a higher number of positively associations were found for analyses using physical activity during adolescence as exposure, life period with higher linear growth. However, lower percentages of positive associations with bone mass measurements were found for physical activity only during adolescence (around one third) and only in young adulthood (almost half) than for physical activity from adolescence to adulthood (around 80%).

Sports practice during adolescence are related to higher physical activity levels in adulthood, so that associations found between physical activity in young adulthood and bone mass could reflect sports activities in the past, which have greater ground reaction force and, therefore, are more osteogenic.[52] This fact would explain why almost all analyses between bone mass and physical activity considering the whole period of adolescence and adulthood were positively associated.

The adjustment for confounders is other aspect that should be appointed. Due to the fact that the body size is highly correlated to bone mass, all studies included weight and/or height or body mass index in the multivariate analyses. Most studies showed only coefficients of linear regressions with adjustment for body size. Thus, it is difficult to know the real differences introduced by the body size. However, the effect of body size could reduce the coefficient of the association between physical activity and bone mass, as observed in study with AGAHLS sample.[46] Calcium intake was not included as confounder in only one manuscript.[43] Other nutritional variables, such as energy intake and phosphorus, protein, carbohydrates, fat, magnesium and alcohol were included in the models of some studies.[48,49,51–54,56,58] Fewer studies considered smoking in their analyses.[48,49,51,52,57,61] Moreover, reproductive factors, such as parity, breastfeeding and time from weaning were included in multivariate analyses of few studies.[48,56,59,60] Since several differences in the statistical tests and adjustment strategies were found in these studies, it is difficult to determine the magnitude of bias that could be introduced by these differences. However, studies about this subject should carefully take into account the whole hierarchical model and its factors in order to avoid biased results.

Although only around one third of analyses between physical activity during adolescence and bone mass measurements were positively associated, when the results by sex are showed, important differences between genders are observed, since the most part of associations were found in males. The lack of association in females, besides biological differences, could be explained by their lower participation in sports and vigorous activities or an insufficient physical activity level to create a demonstrable effect on their bone mass.[26,52] Thus, though participation in moderate activities as walking is not different between genders, in the worldwide context males are more likely to participate in vigorous-intensity physical activity than are females.[62]

In addition, considering differences on effect of physical activity during adolescence on bone mass between genders, it has been suggested that boys' bones are more sensitive to loading than girls' bones.[63] Moreover, it seems that the effect of physical activity on bone status reduces in females, but not in males.[64] However, the most important explanation for lack of association between physical activity and bone mass in females is their less frequency in sports involving high peak strain and ground reaction force enough to increase their bone mass.[26]

From studies included in this review, it is impossible to recommend the amount of physical activity necessary to promote benefits on bone health, since different instruments for physical activity evaluation were used in these studies. In addition, it is impossible to determine the pooled magnitude of effect of physical activity in each age on bone mass due to the same reason. The current guidelines themselves did not report a consistent recommendation for enough physical activity to improve the bone health. Recommendations for children and adolescents only appoint that it is important to spend a percentage of 60 minutes of daily physical activity in bone-strengthening activities on at least 3 days a week. For adults there is no specific recommendation to promote bone health.[65]

It seems to be a consensus that high impact sports are the main activities that maximize bone mass accumulation and maintenance and also reduce the loss of bone mass on elderly and postmenopausal period. However, it is not clear which is the best training method for enhancing bone mass, though scientific evidence points to a combination of high impact exercises with weight-lifting exercises.[5] The studies included in this review did not compare the effect between different activities, but in sample from AGAHLS, associations between physical activity and bone mineral density in lumbar spine and femoral neck in different times of evaluation of both exposure and outcomes were more consistent using peak strain score than when general physical activity was used. This strengthens that current recommendations of physical activity, mainly for adults, may not be adequate to attend the needs of bone health.

Besides type of activities, other difficult questions to be responded by the literature are concerned with how many sessions (frequency) and how long (duration) is needed to cause bone adaptation. Such studies did not respond these questions, but several randomized studies with positive results have used 2–3 training days per week, though this depends on the type of activity practiced.[5]

The pooled findings show that more studies with positive associations between physical activity and bone mass were seen in males than females. The relationship of physical activity only during adolescence or adulthood and bone mass was not found in young women, mostly likely because they did not participate in peak strain activity on a sufficiently frequent basis. Moreover, analyses performed for each period did not discard the effect of physical activity posteriorly or previously and the tracking effect should be considered in this question since people who were highly active in adolescence are more likely to be active in adulthood. In addition to results found in females, since physical activity only during adolescence or adulthood seems to have no effect on bone mass, it is important to promote physical activity in both growth and maintenance periods for them due to the fact that women with more engagement in physical activity in the whole period from adolescence to adulthood may have benefits to bone health as well as males too.

The physical activity during the growth period seems to be highly important for males taking into account the positive effect on total period – from adolescence to adulthood and the maintaining across the lifespan. However, recent publication appointed that few data available indicate that exercise benefits in bone mineral density are eroded in the long term, indicating that residual factors caused by physical activity in the growth period such as structural changes, muscle strength, coordination and balance could be more important to prevent fractures in later life.[66]