Fifty-four million Americans have low bone density or osteoporosis, and about 1 of every 2 women older than 50 years will break a bone because of osteoporosis (31). One recommendation for decreasing the risk of bone fracture later in life is to achieve the highest possible bone mineral density (BMD) by early adulthood. Physical activity during the prepubertal and pubertal period is associated with achieving the highest possible BMD. Evidence of the benefits of physical activity during this time frame is the fact that female collegiate athletes who compete in impact sports typically have higher bone density values than age-matched controls (11,14,22,32,39).
Nearly 90% of peak bone mineral content (BMC) and 96% of BMD achieved in females by 18 years of age (49). What is unknown is whether these collegiate female athletes, who already have high BMD values when they enter college, will significantly increase their BMD during their collegiate careers. Furthermore, there are limited available data on what effects training and participation in different sports will have on further increases in BMD. Only 1 study examining BMD has followed up female collegiate athletes for more than 1 year and this study included only gymnasts (43). In this work, Snow et al. (43) showed a significant longitudinal increase in total body BMD of 2.5% from preseason year-1 to postseason year-2. Longitudinal studies are clearly needed to determine if changes in BMD can be expected in female collegiate athletes from other impact and nonimpact sports and if there are changes, to determine if they differ among different sports.
To achieve the highest BMD possible, 1 consistent recommendation is to participate in physical activities with sufficient bone loading forces (21). However, factors such as the magnitude, frequency, and duration of optimal bone loading forces are unknown (21). In addition, movement specific or novel loading patterns may have differential effects (2,33–35), potentially resulting in bone adaptations unique to particular sports. Several studies have shown that female collegiate athletes in impact sports have higher BMD values than those in a nonimpact sport like swimming (SW) (8,11,14,22,30,48). However, few studies (8,22,30,32) have compared impact sports and only 2 (8,30) have compared more than 2 impact sports in National Collegiate Athletic Association (NCAA) Division 1 female athletes. Carbuhn et al. (8) did not see any statistically significant differences between basketball (BB) and volleyball (VB) or VB and track sprinters and jumpers (TR), with BB being higher than TR for total BMC and leg BMD only. Mudd et al. (30) found no differences among gymnasts, field hockey, soccer (SOC), softball, and TR and that gymnasts were higher than runners. These 2 studies indicate that there are no differences among impact sports; however, additional research is needed to confirm these findings and to extend the comparative database.
Use of whole-body dual energy x-ray absorptiometry body composition and BMD measures to assess collegiate female athletes is becoming more common. The only comparative BMD data generated by the automated DXA reports are for age-matched controls for total BMD only. Because athletes typically have higher BMD values than age-matched controls, more published data for specific sports are needed to assist the medical and training staff in evaluating the individual and team DXA BMD data.
The primary aim of this study was to contribute to this database by assessing the longitudinal changes in BMC and BMD measures in NCAA Division 1 female athletes in impact (BB, SOC, TR, and VB) and nonimpact (SW) sports compared with physically active age-matched controls. It was hypothesized that small but significant increases would be seen in the impact sports. A secondary aim was to compare BMC and BMD measures in impact (BB, SOC, TR, and VB) and nonimpact (SW) sports with physically active age-matched controls. It was hypothesized that (a) the BMC and BMD values of the impact sports would be greater than the nonimpact sport and the controls and (b) there would be little difference between the impact sports.
Experimental Approach to the Problem
This is a companion study to a previously published body composition study (45). The experimental approach, athlete subjects, and procedures used were identical because the bone measures were collected simultaneously with the body composition measures.
To assess cross-sectional and longitudinal changes in female athletes, BMC and BMD measurements were taken at preseason and postseason throughout the collegiate careers of female BB, SOC, SW, TR (sprinters and jumpers only), and VB athletes at a NCAA Division 1 university. Preseason measures were taken in early to mid-August for SOC and VB and in late August to mid-September for BB, SW, and TR. Postseason measurements were conducted in November for SOC, December for VB, February/March for BB and SW, and April/May for TR. Data used for this study were collected over the following time frame: BB and TR from 2003 to 2010, SW from 2006 to 2010, and SOC and VB from 2007 to 2010. All athletes participated in sport-specific aerobic, strength, and power training throughout the year. Basketball training included: September to mid-October: 3 hours of strength training and 2 hours of on-court conditioning or practice each week; mid-October to mid-March: 1.5 hours of strength training along with practices and games each week; mid-March to June: 3 hours of strength training and 2 hours of conditioning and practice each week with 2–3 weeks off at the beginning and ending of this cycle; and June to September: 3 hours of strength training and 2 hours of conditioning and practice each week with 3 weeks off at the end of this cycle. Soccer training included: mid-August to December: 1 day of resistance training and 1 day of conditioning along with practice and games each week; December to mid-January: none to the end of the semester with home workouts during the Christmas break; mid-January to mid-February: 3 days of a combination of conditioning and resistance training and 2 hours of practice each week; mid-February to April: 2 days of conditioning, 3 days of resistance training, and 10 hours of practice each week; April to June: 3 days of a combination of conditioning and resistance training and 2 hours of practice each week with 2–3 weeks off at the beginning of May; and late May to August: 3–4 days of conditioning and 2 days resistance training each week. The TR training included: September to June: 3–4 resistance workouts and 5 days of running/jumping conditioning each week with some tapering in April and May; June to September: those not competing were off and those continuing to compete continued their training. Volleyball training included: August to December: strength training at lower volume and intensity 3 times weekly along with practices and games; mid-December to mid-January: off completely; January to April: 4 days of resistance training and 2 days of conditioning for 6 weeks followed by 6 weeks of 3 days of resistance training and 2 days of conditioning each week along with practices; May: off completely; June to August: monitored resistance training 3 days and conditioning 5 days each week.
Two separate convenience control groups were used for the longitudinal (CON-1) and cross-sectional (CON-2) aspects of the study. These groups included female students from the same university who participated in university physical activity classes twice a week and who had DXA scans during the same period as the athletes. Classes included aerobic dance, walking, weight training, and combined cardiovascular and weight training.
Participants included 212 female athletes, 18–23 years of age, from BB (n = 38), SOC (n = 47), SW (n = 52), TR (n = 49), and VB (n = 26). Their physical characteristics are given in Table 1. All members of BB, SOC, SW, and VB were tested. Most, but not all, TR athletes were tested as determined by the team's athletic trainers. A few scans were missed because of illness, injury, or noncompliance. Because there were more frequent absences for postseason testing during the athlete's final year of eligibility and some athletes did not complete 4 years of eligibility, only 3 years of data were analyzed. CON-1, used in the longitudinal analysis, consisted of 85 female students, 18–24 years of age, who completed 2 DXA scans that were not less than 1 year or more than 3 years apart. CON-2, used in the cross-sectional comparisons, consisted of 170 female students, 18–20 years of age. All participants were informed of the procedures, benefits, and risks before providing written informed consent before each DXA scan for this study, which was approved by the Institutional Review Board at The University of Texas at Austin.
All participants reported to the Fitness Institute of Texas for a whole-body DXA scan. Participants wore light athletic clothing; removed all jewelry, plastic, and metal materials that could affect the X-ray beam; and confirmed they were not pregnant before testing. Time of day, time of last meal, time of last workout, time of last menstrual cycle, or hydration status were not strictly controlled for because of scheduling complexities. Measures included total BMC and BMD, and arm, leg, pelvis, spine, and trunk BMD using a Lunar Prodigy (GE Medical Systems, Madison, WI, USA). Scans were analyzed using enCORE software (version 11.0; GE Medical Systems). Calibration with a spine phantom consisting of calcium hydroxyapatite embedded in a lucite block was performed at least every other day. Calibration never failed and the coefficient of variation was 0.3% for BMD. Height was measured to the nearest 0.3 cm using an adult/infant stadiometer (Perspective Enterprises, Portage, MI, USA) at the initial visit only.
A restricted maximum likelihood linear mixed model (LMM) regression analysis with a compound symmetric heterogenous variance-covariance matrix structure was performed to determine if (a) BMC and BMD values changed over time in the different sports or CON-1 and (b) BMC and BMD values differed among the different sports and CON-2. Analyses were conducted with the PROC MIXED procedure (SAS 9.2; SAS Institute, Cary, NC, USA). The outcome variables of interest were total BMC and total BMD, and arm, leg, pelvis, spine, and trunk BMD. Sport (BB, SOC, SW, TR, and VB), time of season (pre- vs post-), and year relative to first DXA scan were the predictor variables. Basic analytic assumptions were met: data were of normal and equal variance. The main effect of each independent variable and interactions between sport and time of season, and sport and year were determined by the LMM. A final LMM model was then produced after nonsignificant interactions were eliminated. A t-value generated for each comparison was used to test regression coefficients. The empirical cut-off value designating significance for all tests was set to p ≤ 0.05 for all analyses.
Linear mixed models were adjusted for total body mass because it is significantly correlated with BMD (11,13,26,30,32,38,49), which could confound results. It was decided a priori that models would be adjusted for ethnicity (African-American or not African-American), a factor also highly related to bone density (47). Because BMC and BMD may fluctuate during the year (43), 2 different models were created to examine longitudinal changes. In the first model, variables of interest were modeled at the initial time point and at every postseason time point (year [Y]-1 pre, Y-1 post, Y-2 post, and Y-3 post). In the second model, the average of values across each season was used (Y-1, Y-2, and Y-3). Statistically significant differences were designated at the adjusted means.
Reliability measures for all DXA measures were obtained by performing DXA scans 3 times in a single day on a group of athletic (n = 3) and a group of active (n = 6) college-aged individuals. In the athletic/active groups, the coefficient of variation values were 1.1/1.9, 0.9/0.7, 0.6/0.7, 1.3/1.0, 1.5/1.9, 2.4/4.3, and 1.2/1.0% for total BMC and BMD, and arm, leg, pelvis, spine, and trunk BMD, respectively. These values indicate high reliability and match body composition reliability measures previously reported with these same 2 groups (45,47). The Pearson's product moment correlations between total body mass and selected outcome variables for athletes and controls combined were as follows: total BMC (0.77), total BMD (0.44), and arm (0.44), leg (0.41), pelvis (0.42), spine (0.25), and trunk BMD (0.51).
In LMMs, age predicted trunk (p = 0.008) and pelvis (p = 0.009) BMD only. Ethnicity significantly predicted total BMD (p = 0.012) but no other outcomes (p ≥ 0.05). Statistical models retained ethnicity and mass because trimming these factors did not influence other associations. In all tables, the means are adjusted for the following variables: ethnicity, mass, preseason/postseason, sport, test year, and the sport × test year and sport × preseason to postseason interactions. Sport was significantly related to all outcomes (p < 0.001). The sport × year interaction was significant only for BMC (df = 8/532, F = 4.63, p < 0.0001) and trunk BMD (df = 8/526, F = 2.02, p = 0.0418). The sport × preseason to postseason interaction was significant for BMC (df = 1/537, F = 10.48, p = 0.0013), arm BMD (df = 1/579, F = 13.64, p = 0.002), leg BMD (df = 1/548, F = 26.79, p < 0.0001), pelvis BMD (1/547, F = 45.79, p < 0.0001), spine BMD (df = 1/580, F = 11.89, p = 0.006), trunk BMD (df = 1/536, F = 44.54, p < 0.0001), and total BMD (df = 1/571, F = 17.40, p < 0.0001).
Longitudinal changes using the time points Y-1 pre, Y-1 post, Y-2 post, and Y-3 post are given in Table 2. Among the significant changes (p ≤ 0.05) from Y-1 pre to Y-3 post, which are illustrated in Figure 1, were increases of 5.7% in SW, 2.7% in VB, 2.2% in TR, and 1.4% in BB in total BMD and increases of 3.3% in BB, 2.0% in TR, and 1.5% in SOC in total BMC.
Seasonal changes were examined from preseason to postseason (Figure 2) and postseason to the next preseason (data not shown). Notable preseason to postseason changes (p ≤ 0.05) included increases in total BMD only in BB and SW, and total BMC increased only in BB. Leg BMD increased in all sports except SOC. The only significant changes (p ≤ 0.05) from postseason to preseason were decreases in leg, pelvis, and trunk BMD in BB and pelvis BMD in VB and an increase in total BMC for TR.
Adjusted means and comparisons among all sports and CON-2 are shown in Figure 3. The only differences between SW and CON-2 were that SW had higher arm BMD and lower total and spine BMD (p ≤ 0.05). SW and CON-2 had lower values (p ≤ 0.05) than all other sports for most measures. Basketball had significantly higher values (p < 0.05) than all other groups for all measures, with the exception of leg BMD for VB and spine BMD for SOC and TR (p ≥ 0.05). The only significant differences (p ≤ 0.05) among SOC, TR, and VB were that VB was significantly greater than SOC and TR for total BMC and leg BMD, and TR was significantly greater than VB for spine BMD.
The unique aspect of this study was measuring BMC and BMD values across 3 seasons in collegiate female athletes from different sports. Data from athletes from 5 different sports and a physically active control group were collected over a 4- to 8-year period. The ability to track such a large number of athletes over multiple seasons strengthens the power of the analysis and conclusions. Small but significant longitudinal changes were seen in most sports, which is noteworthy considering the relatively high initial BMC and BMD values reported in the impact sport athletes in this study.
In examining longitudinal changes, only 1 other study (43) has investigated a group of female collegiate athletes for a period longer than 1 season. Snow et al. (43) measured gymnasts preseason and postseason over 2 years. Unfortunately, the only common measurement between Snow et al. (43) and the current study is total BMD. Snow et al. (43) showed a significant longitudinal increase of 0.023 g·cm−2 (2.5%) from Y-1 pre to Y-2 post. The current study showed smaller percent increases in total BMD over a similar period: 0.9% in TR, 1.4% in SOC and SW, 1.9% in VB, and 2.1% in BB. However, initial total BMD values in the current study were 13.9–19.1% higher than those reported by Snow et al. (43). In addition, when comparing absolute change, the increases of 0.024 g·cm−2 in VB, 0.028 g·cm−2 in BB, and 0.033 g·cm−2 in SW were as great or greater than the change of 0.023 g·cm−2 reported by Snow et al. (43). That gymnasts increased total BMD is not unexpected considering that the vertical ground forces in gymnastics are potentially 10–18 times body weight (28). This magnitude of force is higher than that reported for other sports: 3–6 times body weight for VB (52), 3.9–4.6 times body weight for net ball (46), a game similar to BB, 7–12 times body weight during triple jumping (37), 6–7 times body weight during long jumping (36), and 4.2 times body weight during sprinting for females (29). Thus, the amount of increase in both BMC and BMD measures in the current study is remarkable, considering the high initial values for the impact sport athletes. It is also of interest to note that many of the increases seen in the current study did not become statistically significant until year 2 or 3, indicating a slow but continual increase over time. Although the changes were less than the 10% increase in peak bone mass needed to reduce the risk of fragility fractures after menopause by 50% (7), it can be argued that any increase in BMD is potentially beneficial, particularly for those with low BMD values. In summary, it seems that many BMC and BMD measures may increase between 1 and 3%, and possibly more, during the collegiate career of female athletes.
Snow et al. (43) also noted a decrease in total BMD from postseason to preseason of 0.3–0.4%. This is contrary to the current study, which found no significant changes in total BMD from postseason to preseason. In addition, very few changes were observed for any variables in the current study from postseason to preseason, most likely reflecting the year-round training status of these athletes. It is unknown what will happen to BMD if and when these athletes stop training, although they will retire from competitive athletics with comparatively high BMD values. Some data indicate that retired female gymnasts retain higher BMD (5), but there is also evidence in male athletes that these differences disappear at an older age (19,20).
Swimming BMD increased significantly from Y-1 pre to Y-3 post by 1.8% for the arms, 1.9% for the legs, 1.0% for the trunk, and 5.7% for total BMD. These changes were similar to those seen in the impact sports. Although these increases are contrary to what might be anticipated from a nonimpact sport, they are in agreement with previously reported BMD increases over the course of a single season in female collegiate swimmers (8). Although the current study was not designed to determine why BMD might increase in a nonimpact sport like SW, 1 possibility is that these athletes performed year-round resistance training. Some resistance training studies in premenopausal women have shown significant increases in BMD (15,24,27,40,44) while others have shown no significant change (1,6,10,12,16,18,23,41,42,50,51). Three of the studies showing a significant increase in BMD were conducted for a year or more (15,24,27), which is in alignment with the current study, where most of the BMD increases seen in SW were not significant until year 3. Finally, Kohrt et al. (21) suggest that bone mass will only increase with overload that is of high enough intensity to increase lean mass. Previous research (45) with this same cohort of SW athletes demonstrated that they significantly increased lean mass by 1.2% from Y-1 to Y-3. Although the current study was not designed to examine causality for these findings, it seems reasonable to speculate that the BMD increases in SW were associated with, and perhaps because of, the weight training and land-based training regimen.
In addition to longitudinal changes, this study examined cross-sectional BMC and BMD values among collegiate female athletes from different sports. The current study confirms the understanding that there are clearly BMC and BMD differences between impact and nonimpact sports (8,11,14,22,30,48). Athletes in the impact sports exhibited remarkably high BMD values. The percentage of athletes with total BMD t-scores ≥1.0 was 100% for BB, 82% for SOC, 93% for TR, and 92% for VB but only 13% for SW. In addition, the average total BMD values for the athletes in the impact sports were 9.7–12.9% higher than for SW. These differences are large enough to be clinically significant (7).
Few studies (8,22,30,32) have examined BMD differences within impact sports. The trends observed in the current study support what may be described as an impact continuum, where BB anchors the high end and SW the low end. Volleyball, TR, and SOC fit in the mid- to upper-range for BMC and BMD variables. These findings are similar to, and consistent with, those of Carbuhn et al. (8); however, that study did not see any statistically significant differences between BB and VB or VB and TR. In addition, their only reported differences between BB and TR were higher total BMC and leg BMD values for BB. It is interesting to note, however, that although not always statistically significant, both studies show BB to be higher than TR and VB for every variable. In addition, both Carbuhn et al. (8) and the current study found VB to be higher than TR for total BMC, arm BMD, and leg BMD but lower for pelvis BMD and spine BMD. Similarly, although not always statistically significant, Lee et al. (22) found BB to be higher than VB and VB higher than SOC for most values. One other study (30) compared bone density measures for SOC and TR in female NCAA Division 1 athletes. In agreement with the current study, Mudd et al. (30) found no significant differences between the 2 sports. Furthermore, the 2 sports were within 3% of each other for all variables in both Mudd et al. (30) and the current study. To summarize these cross-sectional comparisons, BB had higher BMC and BMD values than the other sports. Soccer, TR, and VB were similar, whereas VB seems to have higher total BMC and leg BMD.
Athletes in the impact sports had higher BMD values than SW; however, the SW cohort had fairly normal BMD values based on comparisons to the age-matched control groups and total BMD t-scores. The percentage of SW with total BMD t-scores ≤−1.0, −0.9 to +0.9, and ≥1.0 was 5, 82, and 13%, respectively. For CON-1 and -2, the percentages for the same t-score ranges were 10, 67, and 23%. Although the SW cohort had normal BMD values, 5% had BMD t-scores ≤−1.0, potentially putting them at a higher risk for developing osteoporosis, indicating that these individuals should be referred to the medical staff for further evaluation.
The differences between the impact sports and SW occurred before the beginning of their collegiate careers. The entire time around puberty is one of accelerated bone development (3,4). There is evidence that differences between participants in impact sports and SW are evident during these periods. Elite 7- to 9-year-old female gymnasts had higher weight-adjusted BMD values than elite 7- to 9-year-old female swimmers (9), and competitive females athletes (average age 13 years) in impact sports had higher to a trend for higher BMD than competitive female swimmers (17). Representative of our cohort, previous research on athletes from this same university (25) found that SW specialized at the median age of 10 years and played a median of 2 other organized sports. By comparison, BB, TR, and VB specialized at the median age of 14 years and played a median of 3–4 other organized sports. This finding indicates that SW in the current study probably participated less in impact activities during the puberty period than their impact sport counterparts. Therefore, although not specifically designed to answer this question, the high initial values among those in impact sports may confirm the benefit of impact activities during the puberty period. The lower values seen in SW of college age highlight the need to encourage impact activity for females who specialize in swimming while young.
The primary strength of the current study is that data were collected over a number of years from a variety of different sports. This process increased the number of observations (n > 500) and, consequently, decreased the chances of having data skewed by a single team with unique characteristics. Additionally, the LMM analyses permitted greater statistical power for detecting significant associations. Limitations include (a) lack of information on nutrition and menstrual status, (b) lack of exercise history for the CON groups, (c) lack of control of potential confounding variables, and (d) all subjects were from the same university. In addition, the bone measures from this study were the values collected in the DXA “body composition” mode and not the more clinically relevant lumber spine and dual femur values collected in the “bone density” mode.
In summary, female collegiate athletes can expect to see small, but significant, increases in many BMC and BMD measures over their collegiate careers. In addition, female collegiate athletes in impact sports have much higher BMC and BMD values and total BMD t-scores than SW and female college students, whereas SW and female college students have similar values. Finally, BB tends to have higher BMC and BMD values than SOC, TR, and VB while those sports seem to be similar.
Most female athletes in impact sports have above normal BMC and BMD values. When using whole-body DXA to measure BMC and BMD with collegiate female athletes, athletic trainers and medical personnel should make comparisons with team and sport-specific normative data rather than with values reported for average college females or other sports. Special attention should be given to female swimmers because some of them will have less than optimal BMD values. Further evaluation and strategies to improve BMD may then be warranted for those with the lowest BMC and BMD values.
The authors express their deepest appreciation to the late Tina Bonci, Associate Athletic Director for Sports Medicine, and all of the athletic trainers at The University of Texas at Austin who organized the testing, and the many Fitness Institute of Texas staff members who helped with data collection.
1. Almstedt HC, Canepa JA, Ramirez DA, Shoepe TC. Changes in bone mineral density in response to 24 weeks of resistance training in college-age men and women. J Strength Cond Res 25: 1098–1103, 2011.
2. Bailey CA, Brooke-Wavell K. Optimum frequency of exercise for bone health: Randomised controlled trial of a high-impact unilateral intervention. Bone 46: 1043–1049, 2010.
3. Bailey D, McKay H, Mirwald R, Crocker P, Faulkner R. A six‐year longitudinal study of the relationship of physical activity to bone mineral accrual in growing children: The university of Saskatchewan bone mineral accrual study. J Bone Miner Res 14: 1672–1679, 1999.
4. Bailey DA, Martin AD, McKay HA, Whiting S, Mirwald R. Calcium accretion in girls and boys during puberty: A longitudinal analysis. J Bone Miner Res 15: 2245–2250, 2000.
5. Bass S, Pearce G, Bradney M, Hendrich E, Delmas PD, Harding A, Seeman E. Exercise before puberty may confer residual benefits in bone density in adulthood: Studies in active prepubertal and retired female gymnasts. J Bone Miner Res 13: 500–507, 1998.
6. Blimkie C, Rice S, Webber C, Martin J, Levy D, Gordon C. Effects of resistance training on bone mineral content and density in adolescent females. Can J Physiol Pharmacol 74: 1025–1033, 1996.
7. Bonjour JP, Chevalley T, Rizzoli R, Ferrari S. Gene–environment interactions in the skeletal response to nutrition and exercise during growth. Med Sport Sci 51: 64–80, 2007.
8. Carbuhn AF, Fernandez TE, Bragg AF, Green JS, Crouse SF. Sport and training influence bone and body composition in women collegiate athletes. J Strength Cond Res 24: 1710–1717, 2010.
9. Cassell C, BenedicT M, Specker B. Bone mineral density in elite 7- to 9-yr-old female gymnasts and swimmers. Med Sci Sports Exerc 28: 1243–1246, 1996.
10. Chilibeck P, Calder A, Sale D, Webber C. Twenty weeks of weight training increases lean tissue mass but not bone mineral mass or density in healthy, active young women. Can J Physiol Pharmacol 74: 1180–1185, 1996.
11. Creighton DL, Morgan AL, Boardley D, Brolinson PG. Weight-bearing exercise and markers of bone turnover in female athletes. J Appl Physiol (1985) 90: 565–570, 2001.
12. Dornemann T, McMurray R, Renner J, Anderson J. Effects of high-intensity resistance exercise on bone mineral density and muscle strength of 40-50-year-old women. J Sports Med Phys Fitness 37: 246–251, 1997.
13. Egan E, Reilly T, Giacomoni M, Redmond L, Turner C. Bone mineral density among female sports participants. Bone 38: 227–233, 2006.
14. Fehling PC, Alekel L, Clasey J, Rector A, Stillman RJ. A comparison of bone mineral densities among female athletes in impact loading and active loading sports. Bone 17: 205–210, 1995.
15. Friedlander AL, Genant HK, Sadowsky S, Byl NN, Glüer C-C. A two-year program of aerobics and weight training enhances bone mineral density of young women. J Bone Miner Res 10: 574–585, 1995.
16. Gleeson PB, Protas EJ, Leblanc AD, Schneider VS, Evans HJ. Effects of weight lifting on bone mineral density in premenopausal women. J Bone Miner Res 5: 153–158, 1990.
17. Grimston SK, Willows ND, Hanley DA. Mechanical loading regime and its relationship to bone mineral density in children. Med Sci Sports Exerc 25: 1203–1210, 1993.
18. Heinonen A, Sievänen H, Kannus P, Oja P, Vuori I. Effects of unilateral strength training and detraining on bone mineral mass and estimated mechanical characteristics of the upper limb bones in young women. J Bone Miner Res 11: 490–501, 1996.
19. Karlsson MK, Johnell O, Obrant KJ. Is bone mineral density advantage maintained long-term in previous weight lifters? Calcif Tissue Int 57: 325–328, 1995.
20. Karlsson MK, Linden C, Karlsson C, Johnell O, Obrant K, Seeman E. Exercise during growth and bone mineral density and fractures in old age. Lancet 355: 469–470, 2000.
21. Kohrt WM, Bloomfield SA, Little KD, Nelson ME, Yingling VR. Physical activity and bone health. Med Sci Sports Exerc 36: 1985–1996, 2004.
22. Lee EJ, Long KA, Risser WL, Poindexter HB, Gibbons WE, Goldzieher J. Variations in bone status of contralateral and regional sites in young athletic women. Med Sci Sports Exerc 27: 1354–1361, 1995.
23. Liang M, Braun W, Bassin S, Dutto D, Pontello A, Wong N, Spalding T, Arnaud S. Effect of high-impact aerobics and strength training on BMD in young women aged 20–35 years. Int J Sports Med 32: 100–108, 2011.
24. Lohman T, Going S, Hall M, Ritenbaugh C, Bare L, Hill A, Houtkooper L, Aickin M, Boyden T, Pamenter R. Effects of resistance training on regional and total bone mineral density in premenopausal women: A randomized prospective study. J Bone Miner Res 10: 1015–1024, 1995.
25. Malina RM. Early sport specialization: Roots, effectiveness, risks. Curr Sports Med Rep 9: 364–371, 2010.
26. Mazess RB, Barden HS. Bone density in premenopausal women: Effects of age, dietary intake, physical activity, smoking, and birth-control pills. Am J Clin Nutr 53: 132–142, 1991.
27. McDermott MT, Christensen RS, Lattimer J. The effects of region-specific resistance and aerobic exercises on bone mineral density in premenopausal women. Mil Med 166: 318–321, 2001.
28. McNitt-Gray JL. Kinetics of the lower extremities during drop landings from three heights. J Biomech 26: 1037–1046, 1993.
29. Mero A, Komi PV. Force-, EMG-, and elasticity-velocity relationships at submaximal, maximal and supramaximal running speeds in sprinters. Eur J Appl Physiol Occup Physiol 55: 553–561, 1986.
30. Mudd LM, Fornetti WC, Pivarnik JM. Bone mineral density in collegiate female athletes: Comparisons among sports. J Athl Train 42: 403–408, 2007.
31. National Osteoporosis
Foundation. What is osteoporosis
? Available at: http://nof.org/articles/7
. Accessed July 4, 2014.
32. Nichols DL, Sanborn CF, Bonnick SL, Gench B, Dimarco N. Relationship of regional body composition to bone mineral density in college females. Med Sci Sports Exerc 27: 178–182, 1995.
33. Nikander R, Kannus P, Dastidar P, Hannula M, Harrison L, Cervinka T, Narra NG, Aktour R, Arola T, Eskola H, Soimakallio S, Heinonen A, Hyttinen J, Sievänen H. Targeted exercises against hip fragility. Osteoporos Int 20: 1321–1328, 2009.
34. Nikander R, Sievänen H, Heinonen A, Kannus P. Femoral neck structure in adult female athletes subjected to different loading modalities. J Bone Miner Res 20: 520–528, 2005.
35. Nikander R, Sievänen H, Uusi-Rasi K, Heinonen A, Kannus P. Loading modalities and bone structures at nonweight-bearing upper extremity and weight-bearing lower extremity: A pQCT study of adult female athletes. Bone 39: 886–894, 2006.
36. Ramey MR. Force relationships of the running long jump. Med Sci Sports Exerc 2: 146–151, 1970.
37. Ramey MR, Williams KR. Ground reaction forces in the triple jump. Int J Sport Biomech 1: 233–239, 1985.
38. Reid IR, Plank LD, Evans MC. Fat mass is an important determinant of whole body bone density in premenopausal women but not in men. J Clin Endocrinol Metab 75: 779–782, 1992.
39. Risser WL, Lva EJ, LeBlanc A, Poindexter HBW, Risser JMH, Schneider V. Bone density in eumenorrheic female college athletes. Med Sci Sports Exerc 22: 570–574, 1990.
40. Ryan AS, Ivey FM, Hurlbut DE, Martel GF, Lemmer JT, Sorkin JD, Metter EJ, Fleg JL, Hurley BF. Regional bone mineral density after resistive training in young and older men and women. Scand J Med Sci Sports 14: 16–23, 2004.
41. Sinaki M, Wahner HW, Bergstralh EJ, Hodgson SF, Offord KP, Squires RW, Swee RG, Kao PC. Three-year controlled, randomized trial of the effect of dose-specified loading and strengthening exercises on bone mineral density of spine and femur in nonathletic, physically active women. Bone 19: 233–244, 1996.
42. Singh JA, Schmitz KH, Petit MA. Effect of resistance exercise on bone mineral density in premenopausal women. Joint Bone Spine 76: 273–280, 2009.
43. Snow CM, Williams DP, LaRiviere J, Fuchs RK, Robinson TL. Bone gains and losses follow seasonal training and detraining in gymnasts. Calcif Tissue Int 69: 7–12, 2001.
44. Snow-Harter C, Bouxsein ML, Lewis BT, Carter DR, Marcus R. Effects of resistance and endurance exercise on bone mineral status of young women: A randomized exercise intervention trial. J Bone Miner Res 7: 761–769, 1992.
45. Stanforth PR, Crim BN, Stanforth D, Stults-Kolehmainen MA. Body composition changes among female NCAA division 1 athletes across the competitive season and over a multiyear time frame. J Strength Cond Res 28: 300–307, 2014.
46. Steele JR, Milburn PD. Ground reaction forces in landing in netball. J Hum Mov Stud 13: 399–410, 1987.
47. Stults-Kolehmainen MA, Stanforth PR, Bartholomew JB. Fat in android, trunk, and peripheral regions varies by ethnicity and race in college aged women. Obesity (Silver Spring) 20: 660–665, 2012.
48. Taaffe DR, Snow-Harter C, Connolly DA, Robinson TL, Brown MD, Marcus R. Differential effects of swimming versus weight-bearing activity on bone mineral status of eumenorrheic athletes. J Bone Miner Res 10: 586–593, 1995.
49. Teegarden D, Proulx WR, Martin BR, Zhao J, McCabe GP, Lyle RM, Peacock M, Slemenda C, Johnston CC, Weaver CM. Peak bone mass in young women. J Bone Miner Res 10: 711–715, 1995.
50. Vuori I, Heinonen A, Sievänen H, Kannus P, Pasanen M, Oja P. Effects of unilateral strength training and detraining on bone mineral density and content in young women: A study of mechanical loading and deloading on human bones. Calcif Tissue Int 55: 59–67, 1994.
51. Warren M, Petit MA, Hannan PJ, Schmitz KH. Strength training effects on bone mineral content and density in premenopausal women. Med Sci Sports Exerc 40: 1282–1288, 2008.
52. Wielki C, Dangre M. Analysis of jump during the spike of volleyball. In: Biomechanics IX_B. Winter D., Norman R., Wells R., Hayes K., Patla A., eds. Champaign, IL: Human Kinetics, 1985. pp. 438–442.