Journal Logo


Breast Elevation and Compression Decrease Exercise-Induced Breast Discomfort


Author Information
Medicine & Science in Sports & Exercise: July 2010 - Volume 42 - Issue 7 - p 1333-1338
doi: 10.1249/MSS.0b013e3181ca7fd8
  • Free


Excessive vertical breast displacement during physical activity has been linked to exercise-induced breast discomfort and embarrassment associated with physical appearance (2,4,6-8,15,16,18), both of which are barriers to women participating in physical activity (2,4,6-8,15,16,18). To limit this excessive vertical breast displacement, women are usually encouraged to wear external breast support, such as a sports bra, during physical activity (5,7,8,14,17). Although external breast support is required by a large proportion of the population, particularly those women who wish to participate in physical activity, there is a paucity of published articles on the science of breast biomechanics during physical activity or other such evidence upon which to design effective sports bras. Two main types of sports bras are typically available, encapsulation bras and compression bras. Encapsulation sports bras support each breast in a separate cup and are usually composed of rigid material in the bra cups, strong elastic material in the bra band, and wide padded straps for comfort (7,17). Compression bras, usually known as crop tops, are made of elastic material, which functions to compress both breasts as a single unit against the chest wall. Biomechanical studies have found encapsulation sports bras to be superior to crop tops in limiting vertical breast displacement and exercise-induced breast discomfort (7,17). However, those bras found to be the most effective in reducing vertical breast displacement have also been ranked as the most uncomfortable to wear (7). To promote physical activity and its long-term secondary health benefits for women, further development of sports bra design is warranted on the basis of scientific evidence to ensure breast support options are both functional and comfortable.

Results of a recent biomechanical study of women running in deep water revealed that women with large breasts experienced significantly lower exercise-induced breast discomfort compared with when they ran on a treadmill despite comparable vertical breast displacement during the two running conditions (10). As the women's breasts were significantly elevated by the buoyant forces associated with the aquatic environment during deep water running, it was speculated that the buoyant forces, which counteracted gravitation forces, reduced tension on the passive anatomical breast supports (i.e., the overlying skin and Cooper's ligaments). That is, the upward buoyant forces decreased the instantaneous vertical breast velocity and prevented the breasts from reaching their end range of motion during the downward breast trajectory, reducing "breast slap" (4). However, as synchrony of breast motion relative to the running cycle differed in the two environments, it is unknown whether the reduction in exercise-induced breast discomfort associated with the breast elevation and slower instantaneous vertical breast velocity, characteristic of deep water running, could be replicated during land running through sports bra design modifications.

Previous research has shown that if large ptotic breasts are elevated without simultaneous compression, they protrude further anteriorly from the body in the anterior-posterior direction (3). Such protrusion can increase the breast moment arm relative to the trunk, possibly increasing loading on the breast tissue's anatomical restraints, compromising the support they provide, particularly over the long term via tissue elongation. Therefore, we postulate that when designing a sports bra, breast elevation should be accompanied by breast compression. However, inconsistent results have been reported regarding the effect of additional breast compression worn external to an encapsulation sports bra in terms of reducing exercise-induced breast discomfort (2,17).

The aim of this study was to investigate whether a sports bra designed to both elevate and compress the breasts, in addition to limiting vertical breast displacement, could decrease exercise-induced breast discomfort and bra fit discomfort experienced by women with large breasts relative to a standard encapsulation sports bra that only limited vertical breast displacement. We hypothesized that compared with a commercially available high support encapsulation sports bra, a high support bra that also elevated and compressed the breasts would result in comparable vertical breast displacement, less vertical breast velocity, and reduced exercise-induced breast discomfort and bra discomfort when women with large breasts ran on a treadmill.



Twenty women (mean ± SD: age = 31 ± 8 yr; range = 19-44 yr) who were professionally sized to wear a C+ bra cup were recruited as representative of women with large breasts. All of the subjects were fitted by the one experienced bra fit assessor (D.E.M.), using a professional bra fitting criteria (9). The average reported bra band size of the subjects was size 14 (range size = 10-18; Australian sizing), and their median cup size was DD (range = C-F). As hormone levels can influence connective tissue within the breasts, only subjects who were premenopausal and not currently breast feeding or pregnant were recruited for the study. Furthermore, participants with a history of breast surgery or any musculoskeletal disorder or pain that would inhibit them from running were excluded. All recruiting and testing procedures were approved by the University of Wollongong Human Research Ethics Committee (no. HE05/83), and all subjects gave written informed consent to participate in the study.

Experimental procedures.

Subjects ran on a treadmill in three randomly ordered bra conditions: a commercially available encapsulation sports bra (New Legend sports bra, Berlei; Pacific Brands, Victoria, Australia), an experimental bra that provided both compression and elevation, and a placebo bra. All bras offered a high level of breast support in terms of limiting vertical breast displacement. The experimental bra and the placebo bra both used the same band, underwire, and strap structure as the sports bra, but the rigid material of the sports bra cups was replaced with an elastic rubber material (Dura-Band®; For You Inc., Pittsburgh, PA), which compressed the breasts when the subjects were fitted with the experimental and placebo bra. Two grades of elastic material were used (red and orange of the DURA-BAND® Resistance Exercise Systems;, accessed May 5, 2009), with the material providing greater resistance to stretch used in the larger sized bras (16D+), as previous research has found that vertical breast displacement increases as bra size increases (8). In addition to compression, the experimental bra was designed to provide breast elevation, which was achieved using high-density 1-cm-thick foam pads placed in the inferior or lateral aspect of the experimental bra cup. The size and the shape of the pads changed in accordance with breast mass and functioned to increase the stiffness of the inferior-lateral aspect of the cup to elevate the breast. Therefore, only the experimental bra offered simultaneous breast compression and elevation. All testing was conducted in the Biomechanics Research Laboratory at the University of Wollongong, and subjects wore their own running shorts and shoes during all testing for each bra condition.

During a familiarization period of treadmill running, each subject self-selected a treadmill running velocity, which they could maintain for at least 3 min without experiencing fatigue. As the subjects were selected to represent women from the general population, including those who are currently physically active as well as those who wish to become physically active, there was a wide range of self-selected running velocities (mean ± SD = 8.3 ± 1.3 km·h−1, range = 6.6-12 km·h−1). They then ran on a treadmill (PowerJog GX-100; Expert Fitness UK, Mid Glamorgan, UK) at this velocity in each experimental condition, for three trials of 3-min duration, with at least 5 min of rest between conditions. Data collection occurred after 2 min to ensure steady-state breast motion had been achieved. Immediately before and after the running trials, the subjects were asked to rate their bra fit comfort, breast discomfort, and perceived breast movement using a visual analog scale (VAS) (rated 0-10), with no discomfort or movement rated as "0" and extreme discomfort or movement rated as "10." They were also asked to rate their RPE using the Borg scale (rated 6-20) (13) as well as rank the bras in order of preference to wear during running, considering breast and bra discomfort and perceived breast movement. The bras were ranked 1-3, with 1 being the most preferred and 3 the least preferred. If the preferential ranking for two bras was the same, the marks were split between the two bras.

Breast kinematics.

To quantify breast motion relative to trunk motion, infrared light-emitting diodes (2-mm diameter) were placed directly on both nipples under each bra using double-sided surgical tape, as the nipples have been found to be the best indicator of vertical breast displacement (1,8). Markers were also placed on each subject's sternal notch and left heel to enable later calculation of breast movement relative to trunk motion and heel strike. To ensure that the principal axes of the trunk were aligned to the global coordinate system, makers were also placed on the acromion processes, iliac crests, and spinous process of the 12th thoracic and 5th lumbar vertebra. The three-dimensional position of all markers in space was tracked using two OPTOTRAK 3020 Position Sensors (100 Hz; Northern Digital, Ontario, Canada) positioned 3.6 m away from the subject in the frontal and sagittal plane.

The raw coordinate data for each marker were imported into the Visual 3D Basic software (Version; C-Motion Inc., Germantown, MD), which filtered (fourth-order zero-phase-shift Butterworth digital low-pass filter (fc = 10 Hz)) and smoothed the data. The processed data were then used to derive the following kinematic parameters for a minimum of 30 breast cycles per condition: (i) vertical breast displacement (cm) relative to the torso; (ii) instantaneous vertical breast velocity (cm·s−1); (iii) breast compression (cm), measured as the distance from the nipple to the sternum in the anterior-posterior plane while the subjects were standing still; and (iv) breast elevation (cm), measured by the low point of vertical breast displacement, relative to the sternal notch, during the downward breast trajectory. As the nipples are positioned below the sternal notch marker, breast elevation was expressed as a negative value, whereby increased elevation was represented by a smaller value (i.e., the nipples were elevated and therefore closer to the sternal marker). Left breast kinematics were described relative to left heel strike because of the synchronicity of breast motion and heel strike and because previous research has reported no significant difference between the left and the right breast vertical breast displacement (12).

Statistical analysis.

Mean and SD values were calculated for the average vertical breast displacement, vertical breast velocity, breast elevation during the downward breast trajectory, and breast compression in the static position. The mean VAS scores of the subjective variables characterizing breast and bra discomfort before and after running trials as well as the RPE (Borg scale) were also compared among the three conditions. A repeated-measures ANOVA design (P < 0.05) with one within factor (bra condition) was used to compare the above data across the three bra conditions, with the Bonferroni post hoc analyses used to identify where any significant between-condition differences were. Wilcoxon signed-ranks tests were used to compare the bra rankings. All statistical procedures were conducted using the Statistical Package for the Social Sciences (Version 15.0; SPSS Inc., Chicago, IL).


Vertical breast displacement and velocity.

No significant main effect of bra condition was found on the vertical breast displacement (P = 0.12) or instantaneous vertical breast velocity (P = 0.06) (Table 1). No significant between-condition difference was found in the mean duration of the subjects' strides from left heel strike to left heel strike (P = 0.72) or the mean duration of the downward breast trajectory (P = 0.17).

Mean ± SD vertical breast displacement (VBD) and instantaneous vertical breast velocity (IVBV) of the left breast relative to the trunk when subjects (n = 20) wore the experimental bra, the placebo bra, and the sports bra. There was no significant difference between the three bra conditions.

Breast elevation.

There was a significant main effect of bra condition on breast elevation (P < 0.001). Post hoc analysis confirmed that breast elevation in the experimental bra condition was significantly greater (on average 2 cm greater) than when wearing the sports bra and when subjects wore the placebo bra (Fig. 1).

Mean ± SD breast elevation, signified by vertical displacement of the left nipple (cm) relative to the sternal notch during the downward breast trajectory for the experimental bra (EB), placebo bra (PB), and sports bra (SB). A significant difference was found between thethree bra conditions, with the EB having the greatest breast elevation,signified by the least distance of the nipple below the sternal notch.*Significant between-bra condition difference (P < 0.01).

Breast compression.

There was a significant main effect of bra condition on static breast compression (P < 0.001), whereby breast compression was significantly greater when subjects wore the placebo bra relative to both the sports bra and the experimental bra (Fig. 2). Although no significant difference in breast compression was found between the experimental bra and the sports bra condition (P = 0.27), the experimental bra incorporated a 1-cm-thick pad in the cup that artificially increased this measurement.

Mean ± SD breast compression, signified by the mean distance of the left nipple relative to the manubriosternal angle (sternum) in the anterior-posterior plane for the experimental bra (EB), placebo bra (PB), and sports bra (SB). The PB had significantly more breast compression (P < 0.01) than both the EB and the SB. Although there was no statistical difference between the experimental bra and the sports bra condition for breast compression (P = 0.27), the EB hada 1-cm-thick pad within the bra cup, which would have increased this measurement. *Significant between-bra condition difference (P<0.01).

Breast and bra discomfort.

Before the running trials, the subjects reported no significant between-condition difference in breast or bra discomfort or in the RPE during the running trials (Table 2). However, after running, the subjects reported significantly less breast and bra discomfort and less perceived breast movement when wearing the experimental bra compared with when wearing the placebo bra and the sports bra (Table 2). The subjects reported low VAS scores for exercise-induced breast discomfort for all three bra conditions. The experimental bra was ranked the "most preferential bra to wear during running" (P < 0.05) (Fig. 3).

Mean ± SD VAS scores (0-10) characterizing breast and bra discomfort and perceived movement for the experimental bra, placebo bra, and sports bra before and after the running trials.
Mean ± SD bra ranking ("most preferred bra to wear during physical activity") for the experimental bra (EB), placebo bra (PB), and sports bra (SB). The bras were ranked 1-3, where 1 was ranked the bra most preferred and 3 the bra least preferred. The EB was ranked significantly better than either the PB or the SB. *Significant between-support condition difference (P < 0.05).


Vertical breast displacement.

Consistent with our hypothesis, vertical breast displacement was comparable between all three bra conditions. It was also consistent with previous studies, which have documented comparable magnitudes of vertical breast motion of women running on a treadmill while wearing high levels of breast support (2,7). Previous research has found a strong relationship between vertical breast displacement and exercise-induced breast discomfort (2,7,8,10,17). The design of all three bras used in the current study therefore offered a high level of breast support and, accordingly, relatively low levels of exercise-induced breast discomfort. However, as no significant difference was found in the vertical breast displacement between the three bra conditions, the variations found in breast and bra discomfort in the present study could not be attributed to variations in vertical breast displacement.

Vertical breast velocity.

The instantaneous vertical breast velocity in the present study was consistent with previous research (2,10). We originally speculated that it might be feasible to decrease instantaneous vertical breast velocity by increasing the duration of the downward breast trajectory, as has been found previously in water running research (10). The use of elastic material in the cup aimed to achieve this increased downward breast trajectory duration. However, as the duration of the breast cycle is coupled with stride cadence, it was not possible to affect the within-subject instantaneous vertical breast velocity in the present study, where stride cadence was standardized by ensuring the subjects ran at the same treadmill velocity in all the bra conditions. Therefore, limiting vertical breast displacement appears to be the crucial factor in sports bra design when attempting to minimize instantaneous vertical breast velocity.

Breast elevation.

The subjects' breasts were significantly elevated, particularly during the downward breast trajectory, when they wore the experimental bra. We speculate that this increased breast elevation reduces tension and loading on the anatomical breast support structures because these structures were further from their end of range when the subjects wore the experimental bra compared with when they wore either the placebo or the sports bra. Breast movement is restrained by a combination of the bra and the breast's anatomical supports (the overlying skin and Cooper's ligaments) (4,14), the bra-breast spring. These anatomical structures are passive tissues and would contribute maximally to the bra-breast spring when they are at the end of their lengthened tensile range. Therefore, if these tissues were not at the end of their range, the percentage of their contribution to the bra or breast spring is likely to be less when wearing the experimental bra relative to when wearing either the placebo bra or the sports bra. This may explain why exercise-induced breast discomfort and bra discomfort during the running trials were less in the experimental bra condition compared with the placebo and the sports bra conditions. It is also possible that simply elevating breast mass may decrease breast discomfort by altering the breast's moment arm relative to the postural muscles that control the trunk and the shoulders, although this notion warrants confirmation by further research.

Breast compression.

The significant increase in breast compression when subjects wore the placebo bra compared with the other two bra conditions suggests that incorporating elastic material in the cup can successfully compress the breasts against the torso, even in an encapsulation design. Although there was no significant difference in the measure of compression between the experimental bra and the sports bra, we speculate that the experimental bra applied greater breast compression than the sports bra because it also contained a 1-cm-thick pad in the lower half to one-third of the cup, which would have increased the breast compression measurement. The benefit of this compression, in terms of reducing exercise-induced breast discomfort, was consistent with previous studies that have investigated the compression effects provided by crop tops (11), the compression applied externally and in addition to an encapsulation sports bra (2), and the compression gained by wearing two bras simultaneously during sport (11). Combining the characteristics of encapsulation and compression within the one bra to support the breasts during physical activity instead of wearing two bras is beneficial in terms of increased comfort as well as reduced cost. This is particularly relevant as crop tops, which have no clasps to undo the bra, tend to wear out quickly as they are stretched every time they are put on or taken off.

Breast and bra discomfort.

The significant reduction in VAS scores for exercise-induced breast discomfort, bra discomfort, and perceived breast movement when subjects wore the experimental bra compared with the sports bra and placebo bra confirms the notion that the modified cup design of the experimental bra improved both breast and bra comfort of women with large breasts. The fact that the placebo bra was reported to be less comfortable than the experimental bra and at the same comfort level as the sports bra confirms that breast compression in isolation without breast elevation does not offer substantial improvement in comfort, relative to the rigid cup design of an encapsulation sports bra design. However, combining breast compression with breast elevation achieved a significant improvement in comfort for women with large breasts.

The significant decrease in perceived breast movement when subjects wore the experimental bra contradicts the lack of a between-bra condition difference in vertical breast displacement. That is, the subjects perceived that their breast movement was less when they wore the experimental bra, although there was no significant difference in the vertical breast displacement in each condition. Whether this perception of reduced movement is a result of altering breast motion in planes beyond the vertical or whether it is associated with not allowing the breasts to reach their end range of motion during the downward breast trajectory requires further investigation. The VAS scores of exercise-induced breast discomfort were also very low for all three bras, particularly considering the large breast size of the subjects, suggesting that the level of support offered by three bras conditions was high. The ranking of the experimental bra to wear in preference to the sports bra, despite the very rudimentary construction of the experimental bra prototype, took into consideration breast movement, breast discomfort, and bra fit discomfort. As overall comfort is paramount to encourage women to wear supportive bras during physical activity, this subjective measure is very important.

The participants in the current study were selected to represent "average" adult women with large breasts so that the results of the study would be applicable to "average" women living in the community, including women who were physically active as well as women who were more sedentary but wished to become physically active. Consequently, participant selection criteria were based on age and breast size, and running experience and fitness were not standardized. However, this variation in fitness level and running experience resulted in a wide range of self-selected treadmill running velocities (6-12 km·h−1) that could be maintained by the participants for the required 3 min, without incurring fatigue. Although running velocity may influence breast biomechanics (8), the present study was based on a within-subject design to minimize the effect of this limitation on the study results.


Combining breast elevation and compression in an encapsulation sports bra resulted in significantly lower subjective ratings of exercise-induced breast discomfort, bra discomfort, and perceived breast movement by women with large breasts, relative to when they ran wearing a standard encapsulation sports bra. These results provide scientific evidence upon which to base the design of sports bras, particularly for women with large breasts, so that their breast support options are both functional and comfortable. Such improvements in sports bra design can enable women with large breasts to exercise in greater comfort, allowing them to enjoy the health benefits associated with participating in physical activity.

This grant was funded by a Faculty of Health & Behavioural Sciences, University of Wollongong, research grant, and the results of the present study do not constitute endorsement by the American College of Sports Medicine.

Disclosure statement: This work did not receive funding from the National Institutes of Health, the Wellcome Trust, or the Howard Hughes Medical Institute.

Conflict of interest: There was no conflict of interest.


1. Eden KB, Valiant GA, Lawson L, Himmelsbach J. Three dimensional kinematic evaluation of sport bra design. Med Sci Sports Exerc. 1992;24(5 suppl):S187.
2. Gehlsen G, Albohm M. Evaluation of sports bras. Phys Sportsmed. 1980;8:88-95, 97.
3. Greenbaum AR, Heslop T, Morris J, Dunn KW. An investigation of the suitability of bra fit in women referred for reduction mammaplasty. Br J Plast Surg. 2003;56:230-6.
4. Haycock CE. The breast. In: Shangold M, Mirkin G, editors. Women and Exercise: Physiology and Sports Medicine. Philadelphia: F.A. Davis; 1988. p. 181-5.
5. Haycock CE. How I manage breast problems in athletes. Phys Sportsmed. 1987;15:89-95.
6. James K. Deterrents to active recreation participation: perceptions of year 10 girls. Health Promot J Aust. 1998;8:183-9.
7. Lorentzen D, Lawson L. Selected sports bras: a biomechanical analysis of breast motion while jogging. Phys Sportsmed. 1987;15:128-39.
8. Mason BR, Page KA, Fallon K. An analysis of movement and discomfort of the female breast during exercise and the effects of breast support in three cases. J Sci Med Sport. 1999;2:134-44.
9. McGhee DE, Steele JR, Munro BJ. Sports Bra Fitness. Wollongong (NSW): Breast Research Australia (BRA); 2008. p. 11-2.
10. McGhee DE, Power BM, Steele JR. Does deep water running reduce exercise-induced breast discomfort? Br J Sports Med. 2007;41:879-83.
11. McGhee DE, Steele JR, Munro BJ. Breast support education improves bra knowledge and bra wearing behaviour in young female athletes: a randomised trial. J Physiother. 2010;56:19-24.
12. Mutter E, Geyssant A, Jeannin T, Chaux C, Belli A. Influence of brassiere on breast vertical acceleration during running. In: Proceedings of the 7th Annual Congress of the European College of Sport Science; July, 2002: Athens (Greece): Pashalidis Medical Publisher; 2002. p. 312.
13. O'Sullivan SB. Perceived exertion: a review. Phys Ther. 1984;64:343-6.
14. Page KA, Steele JR. Breast motion and sports brassiere design. Implications for future research. Sports Med. 1999;27:205-11.
15. Robbins LB, Pender NJ, Kazanis AS. Barriers to physical activity perceived by adolescent girls. J Midwifery Womens Health. 2003;48:206-12.
16. Shaw S. Body image among adolescent women: the role of sports and physically active leisure. J Appl Recreat Res. 1991;16:349-67.
17. Starr C, Branson D, Shehab R, Farr C, Ownbey S, Swinney J. Biomechanical analysis of a prototype sports bra. J Textile Appar Technol Manag. 2005;4:1-14.
18. Taylor WC, Yancey AK, Leslie J, et al. Physical activity among African American and Latino middle school girls: consistent beliefs, expectations and experience across two sites. Women Health. 1999;30:67-82.


©2010The American College of Sports Medicine