Musculoskeletal Response to Exercise Is Greatest in Women with Low Initial Values


Medicine & Science in Sports & Exercise:
BASIC SCIENCES: Original Investigations

WINTERS-STONE, K. M., and C. M. SNOW. Musculoskeletal Response to Exercise Is Greatest in Women with Low Initial Values. Med. Sci. Sports Exerc., Vol. 35, No. 10, pp. 1691–1696, 2003.

Introduction: The “initial values” principle of exercise training states those with the lowest initial values of a physiologic system have the greatest capacity for improvement in response to training. We sought to determine whether initial values predicted the musculoskeletal response to training in premenopausal women (N = 31) who participated in a 1-yr program of resistance and jump training designed to improve physical indices of fracture risk. Significant improvements in trochanteric bone mineral density (BMD), hip abductor strength, power, and postural stability occurred in response to training.

Methods: To determine the predictive power of initial values, we performed separate stepwise regression analyses for each variable including the following dependent variables: age, initial value, highest weight lifted during training, and total number of exercise sessions attended.

Results: In each case, the initial value was the most significant predictor of percent change in response to training. Initial values explained 15–29% of the variance in the magnitude of the training response. For each unit lower BMD of the greater trochanter (0.01 g·cm−2), the training response was 12% greater. For each unit decrease in initial strength (1 N·m), power (1 W), and stability (1 SI unit), the training response was 1.0%, 0.2%, and 8.0% greater, respectively. When categorized by quartile of initial values, women in the lowest quartile had two- to fivefold greater improvements in musculoskeletal measures than those in the upper quartile.

Conclusion: Women who began training with the lowest initial values had the greatest improvements in hip BMD, hip abductor strength, leg power, and postural stability. These results support the training principle of initial values and suggest that this training program may be most successful in premenopausal women with lower values of musculoskeletal indices of fracture risk.

Osteoporosis-related fractures commonly occur as the result of both skeletal frailty and falls. Impact exercise may provide an inexpensive, time-efficient, and effective strategy for increasing bone mineral density (BMD) in premenopausal years as a means to prevent or delay the onset of osteoporosis (3,12,23). The addition of resistance exercise may lower the risk of falls through improvements in lower-body strength, power, and stability—physical factors related to falls (18,19). Although there are numerous reports that exercise improves the musculoskeletal system, individual responses are highly variable and likely depend upon initial values as well as the level of effort and length of the training program. However, the contribution of initial values and exercise stimulus to the training response for the musculoskeletal factors related to fractures is poorly understood.

The initial values principle of exercise training states that individuals with lower initial values of a physiologic variable will exhibit the greatest improvement in response to specific exercise training; however, this principle has only been scientifically evaluated for some physiologic systems. For example, individuals with higher resting heart rate or systolic blood pressure show the greatest reductions in these variables from exercise training compared with individuals who began training with lower initial values (7). However, baseline values fail to explain the variability in high-density lipoprotein (HDL)-C reductions or V̇O2max improvements in response to endurance training, suggesting that determinants of training responsiveness may be complex and dependent upon the particular system being stressed (5,6,16,17). Despite a general lack of supportive evidence, the principle of initial values remains an enduring tenant of exercise training in texts and lay literature.

Based on the principle of initial values, the ACSM Position Stand on osteoporosis suggests that individuals with lower initial BMD may exhibit greater responses to osteogenic exercise (1). Though not advanced in the Position Stand, it may be suggestive that individuals with a greater propensity to fall would also exhibit greater reductions in fall risk as a result of exercise training targeted at improving lower body strength, power, and balance (11). A comprehensive exercise-training program aimed to improve BMD and lower fall risk may therefore most benefit those with a higher pretraining fracture risk profile. To better understand this phenomenon in premenopausal women, we reevaluated data from our intervention study in which we observed significant musculoskeletal improvements within the exercise group that were different from controls (23). Women in the original exercise group increased trochanteric BMD (2.5% ± 2.7%), leg strength (10.9% ± 11.4 and 23.4% ± 28.0 for knee extensor and hip abductor strength, respectively), leg power (25.9% ± 20.9), and stability (24.7% ± 23.5). Our aims were: 1) to determine whether initial values of hip BMD, strength, power, or postural stability would emerge as the best predictor of the respective training response; and 2) to compare the magnitude of the training response between women in the lowest and highest quartiles of initial values.

Author Information

Bone Research Laboratory, Department of Exercise Science, Oregon State University, Corvallis, OR

Address for correspondence: Kerri M. Winters-Stone, Ph.D., School of Nursing, Mailcode: SN-ORD, Oregon Health & Science University, Portland, OR 97239; E-mail:

Submitted for publication March 2003.

Accepted for publication June 2003.

©2003The American College of Sports Medicine