Identifying effective design features of commercial sports bras

Abstract

This paper identifies the effective design features of commercial sports bras by evaluating the reduction of breast displacement during activities. Four subjects, with breast and body types representative of Chinese females of age 24–40 years were studied while they performed activities either braless or wearing one of seven different sports bras representative of the range commercially available. The three-dimensional movement of the breasts was derived by videoing the motion of breast markers attached at six different breast positions using a Vicon 3D motion analysis system. The breast displacement relative to the thorax was measured using a validated local “thorax–breast 3D coordinate system”. The results showed that there was a significant difference in breast displacement between the different breast markers. The bra samples generally achieved the greatest reduction in breast displacement in the medial–lateral direction. The reduction in breast displacement was positively related to the gore height, shoulder strap width, neckline height and side seam depth. The underlying mechanism of breast movement and bra support is also discussed. The study concluded that the most effective bras had the following features: compression type, short vest style, high neckline, slings, cross back, bound neckline, no centre gore, no wire, no cradle, no pad and a non-adjustable wide strap. This provides practical guidelines for bra designers and manufacturers to develop more supportive bras for women performing these physical activities.

Keywords

three-dimensional breast movement bra design coordinate system

Previous research has shown that sports bras are effective in controlling breast movement during running exercises.¹ However, the design and fit of the bras is critical. Inappropriately designed and/or ill-fitting bras with insufficient breast support can lead to musculoskeletal pain, upper limb neural symptoms, deep bra furrows and discomfort.¹ Sports bra designers invariably use stiff fabric around the breasts together with tight shoulder straps to minimize excessive breast movement. However, there is little published information on the critical features^2–7 and the interaction of the components within a sports bra and how to improve the design of sports bras to control breast movement.

To improve the functional design of running sports bras, a scientific study of 3D breast movement for different breast sizes and shapes is necessary. Previous studies^2,8–21 have investigated the effectiveness of bras made from different materials with different styles. However, design rules and the scientific methodology to establish the design criteria are yet to be proposed. In this respect, a standard and valid breast local coordinate system for measuring the 3D breast movement relative to the thorax is required to enable comparative evaluations to be made. This has been, hitherto, an unexplored area in respect of sports bra design. Therefore, the aim of the current research was to determine the functional design features of more supportive sports bras. The fundamental research questions were as follows.

Q1. How do the breasts move, relative to the thorax in 3D, under braless and sports bra conditions during different activities?

Q2. What are the effective features of sports bras for controlling breast movement?

Q3. How can the design of sports bras be improved to reduce the breast movement for the wearer when performing activities such as walking, running and stepping?

To answer these research questions, a 3D motion study using a 3D local coordinate system was carried out to investigate systematically the bra factors that affect breast movement during activities. The objectives were to investigate the 3D breast movement under different support forces that provided 10 specific bra components (Figure 1) during various activities and to evaluate the effectiveness of different bra features on the reduction in breast movement.

Figure 1.

Directions of support force provided by ten bra components: (A) rigid gore, (B) elastic cup, (C) semi-elastic neckline, (D) semi-elastic underarm, (E) rigid wire, (F) rigid cradle, (G) elastic wing, (H) elastic shoulder strap, (I) elastic underband and (J) rigid fastener.

Breast motion studies started in the 1970s. The breast displacement,^{2,5–7,11–14,19,21–24} velocity,^17,19,21,22 acceleration,^7,21,23 and movement trajectory^8,9,12,13,22 have been investigated for subjects in either braless or wearing different styles of bras and performing different activities.^{2,5,7–11,13,17–19,21,22,24–30}

Sports bras can be classified as being either compression or encapsulation types.^14,31–33 The compression bra is designed to restrict breast movement by flattening the breasts against the chest wall, while the encapsulation bra contains two moulded cups which support two separated breasts.³¹ The compression bras have generally been thought to be more effective for women with smaller breasts (cup sizes A or B), while the encapsulation bras were considered to be more effective for women with cup size C or above.³² However, White et al.¹⁸ found no significant differences between these two types of sports bras in controlling breast movement.

Previous researchers found that there were significant differences in breast movement between different bra designs.^11,27 There was no significant difference in the vertical displacement between the left and right breasts¹⁷ and, not surprisingly, breast displacement during running was greater than that during walking.

Most studies have focused only on the vertical movement of the breast,^{2,6–8,13,14,22} because it has been reported to be greater than the horizontal one.^10,12,13 Researchers have tended to use the nipple to represent the whole breast.^{2,5–7,13,14,22,23} However, as the breast is a visco-elastic deformable body whose movement in 3D space is complex, the motion of only one breast point in the vertical direction is probably insufficient to describe the whole breast movement. Therefore, a further objective in the design of this study was to test the following two hypotheses (H₁ and H₂).

H₁: There is a significant difference between nipple displacement and the displacement of different breast quadrants.

H₂: There is a significant difference in breast displacement while walking, running and stepping.

The key obstacle for meaningful biomechanical research is the absence of a reliable local coordinate system to describe breast motion.³⁴ Throughout the previous studies, breast movement relative to the thorax was not measured accurately. In this study, breast movement relative to the thorax was measured by using a newly developed breast coordinate system (BCS) as shown in Figure 2.³⁴ The system was used to define the 3D displacement of six markers on different breast quadrants of four woman subjects wearing different sports bras or no bra while they performed three different activities, namely walking, running and stepping. The results were analysed statistically to determine the effectiveness of bras in reducing breast displacement. From this, the design features within each bra sample that would be most relevant to the development of the ideal sports bra were inferred.

Figure 2.

Positions of markers in a BCS.

Experimental details

Subjects

Four healthy Chinese female casual runners aged from 24 to 40 years with body mass indices ranging from 18.7 to 23.6 kg/m² volunteered to participate in this study. Prior ethical approval was obtained. The procedures to be employed were carefully explained to the subjects before they signed the informed consents to participate. The inclusion criteria were healthy and premenopausal women. To avoid the influence of abnormal hormonal conditions on the connective breast tissues,¹⁷ subjects were excluded if they were breast feeding or pregnant. Those with a history of previous breast surgery or any musculoskeletal disorder or pain were also excluded because that would inhibit their activities. The subjects were selected because they had the most prevalent breast sizes in China, 75B, 75 C, 80B and 80 C, according to a breast sizing survey of 456 Chinese women conducted by Zheng.³⁵ To control the confounding factors of breast size change, such as hydration level and nutritional intake, the subjects were asked to participate in the morning after breakfast. The measurements and characteristics of their breasts are given in Table 1. The breast mass was calculated by multiplying the breast volume by a constant density, 1017 kg/m³.³² The breast volume was measured by a 3D body scanner (TC², US).

Table 1.

Breast characteristics and measurements of the subjects

Subject	Breast size	Bust girth (mm)	Underbust girth (mm)	Breast shape	Breast mass (g)	Body morphology
1	75B	860	740	Hemi-sphere	254	Ectomorphic
2	75 C	915	760	Conical	305	Ectomorphic
3	80B	935	805	Tear	316	Mesomorphic
4	80 C	940	790	Tear	335	Endomorphic

Bra samples and fitting procedures

Seven types of running sports bras comprising two compression bras and five encapsulation bras were tested in this study. The bras were selected as being representative of the different features of sports bras currently commercially available. The bra samples were manufactured under the brand name Shock Absorber™ (DBApparel, UK). The subjects’ breast sizes were measured by a bra fitter who had been trained for 14 weeks in a bra fitting course in the University. The most appropriate bra size was provided for the subject to try on, according to her breast size based on the Metric Bra Sizing System,³⁵ which is determined by the full bust girth and underbust girth (cm).

The fitting procedure follows a standard process. Firstly, the correct tensions of the underband and the shoulder straps were ensured. Secondly, the subject leaned forward so that her breasts completely filled the cup, to ensure that there was no gap, bulging, wrinkle, digging or sliding. Lastly, she raised her hands to check that the bra stayed in place. The bra fit was then assessed and ensured by the same bra fitter, according to a professional bra fitting checklist (Table 2) with detailed criteria in ten bra areas (Figure 1).

Table 2.

Bra fitting checklist

(A) Gore ○ Gore sits against the sternum ○ Allows comfortable breathing ○ Gore width fit for the purpose ○ Wire tip does not dig into the flesh (B) Cup ○ No gap inside the cup ○ Cup seam or lining does not itch ○ No wrinkle on the surface ○ Breast is fully enclosed ○ Smooth shape ○ Cup capacity is sufficient (C) Neckline ○ No gap ○ No bulging ○ Symmetric and balanced ○ Thin, soft and smooth (D) Underarm ○ No gap ○ No extra fabric ○ No bulging ○ Not too much pressure (E) Wire, if any ○ Wire matches breast root ○ Correct gauge ○ Does not dig into the flesh ○ Correct size and width

(F) Cradle ○ Keeps the breast inside the cups ○ Does not curl up (G) Wing ○ Level around the body ○ Appropriate tension to hold the bra in place (H) Shoulder strap ○ Correct tension for breast support ○ Allows enough adjustment ○ Strap does not easily fall off ○ No cutting in ○ No fatigue (I) Underband ○ Tension allows comfort breathing ○ No riding up during motion ○ Bra sits securely while arms are raised (J) Fastener ○ Hooks and eyes wide enough for the style and size

This study focused on the evaluation of the functional performance of the bras in terms of the reduction of breast displacement during activities. The aesthetic, tactile, thermal, moisture and convenience comfort values were assumed to be acceptable. A total of seven styles of running sports bras with different fibre contents and design features were selected for 3D motion analysis. The variables in the design features of the selected bra samples are shown in Table 3.

Table 3.

Differences between M4 and other markers in breast displacements

Direction	Markers in comparison	Mean difference (mm)	Significance (p-value)
x	M4–M1	2.27	0.000*
	M4–M2	1.37	0.000*
	M4–M3	1.68	0.000*
	M4–M5	1.91	0.104
	M4–M6	2.56	0.018*
y	M4–M1	1.59	0.000*
	M4–M2	0.84	0.002*
	M4–M3	1.02	0.657
	M4–M5	1.31	0.007*
	M4–M6	2.22	0.001*
z	M4–M1	1.57	0.000*
	M4–M2	1.69	0.000*
	M4–M3	1.04	0.000*
	M4–M5	1.40	0.615
	M4–M6	2.10	0.154

M1: inner breast, M2: bottom breast, M3: outer breast, M4: nipple, M5: upper breast, M6: top breast. *p-value < 0.05.

The detailed characteristics of the combined variables of the bra samples are shown in Table 4. In the bra industry, the core bra size is 75B which fits wearers of 75 ± 2.5 cm underbust and 87.5 ± 2.5 cm full bust girth, with 12.5 ± 1.25 cm differences between underbust and full bust girths (http://www.wacoal.com.hk). The bra samples in the core size were carefully measured by the same bra fitter using a calibrated tape measure. The definitions of the measurements relevant to bra support are shown in Figure 3, and the variables of the bra measurements are shown in Table 5.

Figure 3.

Definition of bra measurement.

Figure 4.

RBD at six different markers in three different directions.

Table 4.

Detailed descriptions of sports bra samples

Bra sample number		1	2	3	4	5	6	7
Product sketches	Front
Product sketches	Back
Design features	Cup layers	2	2	1	1	2	1	1
	Front cup	Full cup	Short vest	Short vest	Full cup	Short vest	Full cup	Full cup
	Cup seam elongation (%)	30	40	5	Mould	30	8	Mould
	Inner lining	Mould	Mould	No	No	Mould	No	No
	Sling	Yes	No	Yes	Yes	No	Yes	Yes
	Neckline elongation (%)	35	34	14	14	5	20	30
	Gore	Yes	No	No	No	No	Yes	Yes
	Wire	No	No	No	Yes	No	No	Yes
	Cradle	Yes	No	No	Yes	No	Yes	Yes
	Back design	Racer-back	Racer-back	Racer-back	Racer-back	Racer-back	U-back	Cross-over
	Closure	Side	Back	Side	Front	Side	Back	Side
	Shoulder strap elongation (%)	35	50	28	35	30	10	10
Fibre contents	Polyamide (%)	42	55	18	67	5	56
	Polyester (%)	53	16	91	72	28	85	27
	Elastane (%)	5	9	10	10	10	15
	Cotton (%)	29	5	2
Design types		Encapsulation	Compression	Compression	Encapsulation	Compression	Encapsulation	Encapsulation
Claimed support level		Full	Medium	Full	Light	Medium	Full	Full

Table 5.

Measurement of bra samples for core size 75B

Bra style	Neckline to Bust point (mm)	Shoulder width (mm)	Underband width (mm)	Side depth (mm)	Centre front height (mm)
1	60	30	25	90	60
2	65	30	25	85	115
3	85	25	25	95	170
4	60	25	25	75	00
5	55	28	25	85	70
6	75	20	25	100	75
7	70	12	25	100	55

To obtain an assessment of the elasticity of the assembled bra components, such as cup seam, neckline and shoulder straps, the procedure used was to hold the end of the component in one hand and pull the other end by hand to its maximum extension. The elongation of the component was defined as the percentage of the maximum manual extension relative to its original length. The limitations of this technique are acknowledged, but it was easier and more convenient to evaluate this manually rather than to use a tensile testing machine because of the interaction of the components within the bra. Furthermore, this emulated what a potential user would invariably do when selecting a bra for purchase.

Various bra sizes from 75B to 80 C were provided for the subjects to ensure the best fit on their bodies before the motion experiments, according to the afore-mentioned fitting procedure and bra fitting checklist (Table 2).

The aim of this study was to identify the bra(s) that gave the best support to the wearer during the physical activities and from this determine the bra features that are critical. The support functionality of a bra embraces both the design/style of the bra and the material from which it is constructed, and there is a clear interdependency between them. It was beyond the scope of this investigation to study the separate effects of the fabric constructions, their properties and the design of the bras themselves. This study was a prelude to subsequent studies to investigate more systematically the effects of these as independent variables.

Motion analysis

The breast motion analysis was carried out in a Human Locomotion Laboratory under a controlled temperature of 23 ± 0.5 ℃ and relative humidity of 65 ± 3%. As the breast size and its tenderness may slightly change before and during the menses, the experiment for each subject was completed in a single day within seven days after her menses flow. The experiment was conducted in the autumn season and the controlled experimental condition eliminated the variation of sweating that might slightly affect the friction of the bra material. Spherical retro-reflective markers (9.5 mm in diameter, 1.81 g in mass) were attached to the subject’s breast skin or to the sports bras using double-sided adhesive tapes. While the subject was walking, running or stepping, the 3D coordinates of the markers were recorded at a 120 Hz sampling frequency using a Vicon motion analysis system (Vicon 612, Oxford Metrics, Oxford, UK). The system comprised six infrared cameras mounted on a 2.8 m high ceiling in a 102 m² room, and it was statically and dynamically calibrated before the experiments. In this study, passive markers were selected because these wireless markers were preferred to avoid breast deformation under the bra.

Four breast boundary markers were used to define the BCS, as shown in Figure 2. They were named BR (nipple on the right breast), BL (nipple on the left breast), BI (most medial point on the left breast intersecting with the breast root and the transverse plane through BR and BL) and BO (most lateral point on the left breast intersecting with the breast root and the transverse plane through BR and BL). Four reference markers at IJ (deepest point of the incisura jugularis (suprasternal notch)), PX (processus xiphoideus, the most caudal point on the sternum), C7 (spinous process of the 7th cervical vertebra) and T8 (spinous process of the 8th thoracic vertebra) were used to define the thorax. The four breast boundary markers and four reference markers were captured at the same time while the subject was standing upright. Then the boundary markers were removed, and the four reference markers and the six breast markers were captured by the Vicon motion analysis system in a global coordinate system (GCS).

For comparison with previous studies of the nipple movement,^{5,8,14–18,22} the nipple point was selected as one of the marker positions. As the breast movement was non-linear in nature, the nipple movement may not have represented the entire breast motion. Therefore, five more points on the breasts were used to describe the complex pattern of breast movements.³⁶ The reasons and methods used for selecting the marker positions are explained as follows.

When the breast is divided into four quadrants, it can be seen that each quadrant varies in shape and volume, depending on the spatial distribution of fat tissues and skin thickness.³⁷ The upper part of the breast is thinner than that in the adjacent chest wall and axilla regions.³⁸ A previous investigation suggested that the skin in the lower breast region is the thickest (1.94 ± 0.36 mm), and that in the upper region is the thinnest (1.32 ± 0.27 mm), whereas the outer region is 1.62 ± 0.26 mm and the inner region is 1.41 ± 0.24 mm.³⁹ The upper breast has a larger parenchyma ratio over fat tissue. It also possesses Cooper's suspensory ligaments, which help to support the breast weight and prevent sagging.⁴⁰ The subjects’ breast sizes ranged from 75B to 80 C. The surface measurements from the “inner-breast point to the nipple” and “the outer-breast point to the nipple” of the breast size 75B were 81 mm and 101 mm, respectively, while those of the breast size 80 C were 85 mm and 108 mm respectively. The mid height of the “breast cone” was approximately at a surface distance of 40 mm from the nipple, where the markers were located consistently for all the subjects.

As six experimental markers were chosen to attach onto the left breast (or over each well-fitted sports bra), the first marker M4 was placed on the nipple. The markers M1, M2, M3 and M5 were 40 mm apart from M4 in the horizontal and vertical directions, and M6 at the top part of the breast was positioned 40 mm above M5 because the vertical breast displacement was known to be larger than the horizontal one (Figure 2). The 3D coordinates of these four reference markers and six experimental markers were recorded while the subject was stationary with reference to the GCS.

The subjects performed three different activities – walking at 3 km/h and running at 7 km/h on a treadmill, and stepping up and down on a platform 240 mm high for one minute. Each subject preliminarily familiarized herself with her own stepping speed, which was timed and controlled by a metronome to ensure that the subjects stepped at a consistent speed in each experiment. The subjects were allowed to rest for 5 minutes between each experiment.

Data analysis

The 3D coordinates of the six experimental breast markers were first smoothed, using a low pass filter with a cut off frequency of 8 Hz. To eliminate the movement of body rotation, translation and tilting during activities, the data were then transformed from the GCS to the BCS, using a translation and rotation matrix.⁴¹ as shown in equation (1)

[x' y' z'] = [x'_{1} x'_{2} x'_{3} y'_{1} y'_{2} y'_{3} z'_{1} z'_{2} z'_{3}] [x_{1} - x_{0} x_{2} - x_{0} x_{3} - x_{0} y_{1} - y_{0} y_{2} - y_{0} y_{3} - y_{0} z_{1} - z_{0} z_{2} - z_{0} z_{3} - z_{0}]^{- 1} \times [x - x_{0} y - y_{0} z - z_{0}]

(1)

where (x₀, y₀, z₀) is the coordinate of the BCS origin in the GCS, (x₁, y₁, z₁), (x₂, y₂, z₂) and (x₃, y₃, z₃) are any three nonlinear reference coordinates in the GCS; (x′₁, y′₁, z′₁), (x′₂, y′_2, z′₂) and (x′₃, y′₃, z′₃) are their transformed coordinates in the BCS and (x, y, z) and (x′, y′, z′) are the coordinates of the investigating marker in the GCS and BCS, respectively.

Then, the values of the positive and negative peaks within the motion plots for every movement cycle were obtained for each subject, breast point, direction, activities and wearing states. The breast displacement in every movement cycle is defined by the peak-to-peak amplitude of displacement (i.e. positive peak displacement minus negative peak displacement). The average breast displacement in every single experiment was determined by taking the average values of breast displacement obtained in 10 successive movement cycles. The transformation of 3D coordinates from the GCS to the BCS was verified by static and jogging experiments for the same four subjects and by comparing the distances between any two markers in the GCS and BCS under different wearing and activity conditions. The BCS had previously proven to have a very high reliability (ICC = 1.00) for 11 subjects standing at three different positions, in both braless and sports bra conditions, in a static state.⁴²

The breast displacements obtained from the different markers were compared with that of M4 (nipple marker) by paired t-tests and the significant differences were checked at a level of 95%. Analysis of variance (ANOVA) with an LSD post-hoc test was used to test the significant difference of breast displacement during different activities, using SPSS software (SPSS Inc., Illinois, USA).

To measure how well a bra reduced the breast displacement, a unique performance indicator, “Reduced percentage of Breast Displacement” (RBD), was devised. It is defined as the percentage change of displacement (i.e. breast displacement (braless) minus breast displacement wearing a bra, divided by breast displacement (braless) × 100). It follows that the larger the RBD is, the better is the performance of the sports bra.

Results

Comparing different markers

The significance of the difference in the average peak-to-peak amplitude of displacement between M4 (nipple) and each marker was determined by the paired t-tests in x, y and z directions. There were 96 data sets comprising four subjects, three activities and eight bra-wearing states (seven sports bras and one braless). The mean difference and the values of significance are shown in Table 6.

Table 6.

Difference between the nipple marker and other markers in mean breast displacement averaged across four subjects, three activities and eight bra-wearing conditions (n = 96) in three displacement directions

Direction	Markers in comparison	Mean difference (mm)	Significance (p-value)
x	M4–M1	2.72	<0.001*
	M4–M2	1.86	<0.001*
	M4–M3	1.63	<0.001*
	M4–M5	2.39	0.104
	M4–M6	3.96	0.018*
y	M4–M1	2.76	<0.001*
	M4–M2	1.26	0.002*
	M4–M3	1.38	0.657
	M4–M5	2.17	0.007*
	M4–M6	4.01	0.001*
z	M4–M1	1.52	<0.001*
	M4–M2	1.58	<0.001*
	M4–M3	1.46	<0.001*
	M4–M5	2.19	0.615
	M4–M6	3.99	0.154

M1: inner breast, M2: bottom breast, M3: outer breast, M4: nipple, M5: upper breast, M6: top breast. *p-value < 0.05.

In 11 out of 15 cases, the displacements at the five additional markers were different from those of M4 (nipple) at a significance level of 95%. Therefore, H1 was not rejected. In particular, M1 (inner breast) and M2 (bottom breast) had significant differences in breast displacement compared with M4 in all three directions. M6 (top breast) had the largest and most significant mean difference in breast displacement, up to 2.56 mm, with the nipple in both x and y directions. The results imply that measuring only the nipple displacement was insufficient for a scientific analysis of detailed breast movement for the design of supportive bras for subjects performing these activities. Table 6 also shows that M5 (upper breast) did not show a significant difference with M4 in both the x and z directions. Therefore, M5 may be eliminated in future studies, and five measuring points would be optimal for comprehensive breast motion studies.

Comparing different activities

The breast displacement during different activities was compared using ANOVA, to determine the activity with the highest breast displacement at a 95% significance level. There were 32 sets of data of breast displacement from the four subjects for eight bra states. Table 7 shows the p-values obtained and the values in brackets present the mean breast displacement in the x, y and z directions.

Table 7.

ANOVA results of mean breast displacement averaged over six breast markers for four subjects and eight bra conditions (n = 32) in x, y and z directions during different activities (*indicates significant difference with p < 0.05)

Direction	Activity	Activity in comparison	Significance (p-value)
x	Walking (5.19 mm)	Running (8.47 mm)	0.004*
	Walking (5.19 mm)	Stepping (8.33 mm)	0.005*
	Running (8.47 mm)	Stepping (8.33 mm)	0.904
y	Walking (4.31 mm)	Running (9.25 mm)	0.021*
	Walking (4.31 mm)	Stepping (10.42 mm)	0.004*
	Running (9.25 mm)	Stepping (10.42 mm)	0.577
z	Stepping (8.33 mm)	Walking (5.00 mm)	0.074
	Stepping (8.33 mm)	Running (11.00 mm)	0.091
	Walking (5.00 mm)	Running (11.00 mm)	0.001*

In the anterior–posterior (x) direction, the mean breast displacement during walking (5.19 mm) was significantly smaller than those during running (8.47 mm, p = 0.004) and stepping (8.33 mm, p = 0.005). The same trend was evident in the vertical (y) direction; the mean breast displacement during walking (4.31 mm) was significantly smaller than that during running (9.25 mm, p = 0.021) and stepping (10.42 mm, p = 0.004). However, in the medial–lateral (z) direction, the mean breast displacement during stepping (7.79 mm) showed no significant difference when compared to the mean values for walking (4.74 mm, p = 0.074) and running (10.76 mm, p = 0.091). Therefore, H2 was not rejected, except for the medial–lateral displacement.

Reduction in breast displacement (RBD %) at different breast markers

The RBD for the four subjects, performing three activities wearing the seven bra types, were averaged. Figure 4 shows the mean RBDs (n = 84) at the different markers in the different directions. It was found that all the bras generally exhibited the strongest RBD in the medial–lateral (z) direction. In contrast, the RBD was the smallest in the anterior–posterior (x) direction. Of all the six different markers, the RBD at M6 (the top part of the breast) was the lowest.

The bra samples were most effective in controlling the breast movement in the medial–lateral direction, but less effective in the anterior–posterior direction, so it is evident that more attention should be given to the design and construction of the centre front, the plasticity of the cup fabric and the location of the cup seam and neckline to address this. For some bra styles, the poor control at the top part of the breast was attributed to a gap between the bra and the breast inside the top cup during activity, so it should be ensured that the neckline height fits the upper breast boundary, not just while standing, but also during exercise.

Comparison of RBD during different activities

Figure 5 shows the box plots (n = 168) of the average RBDs of seven different bras averaged across four subjects and six markers types in the three directions during the three different activities. The top, middle and bottom lines of the box represent the 75th, 50th and 25th percentiles, respectively, while the top and bottom ends of each vertical line show the maximum and minimum values, respectively. The variance among the different bra samples, subjects and markers can be analysed from the range of box plots.

Figure 5.

Box plots of RBDs during different activities.

Looking at the ranges of RBD, it can be seen that the smallest ranges were during running in the y direction (range = 16%) and during walking in the z direction (range = 19%). This implies similar performances for the different bras in reducing the vertical breast displacement during running, and reducing the medial–lateral breast displacement during walking. Consequently, the design features of the bras are less critical in terms of reducing movement in these directions, so more attention should be paid to the anterior–posterior (x) breast displacement. The median RBDs in all directions during running and stepping are very close to each other. The medians are smaller in the anterior–posterior (x) direction (56% to 58%), but the ranges are wide (29% to 74%).

Comparison of RBD for the different bra styles

Figure 6 shows the mean RBDs for the seven different bra styles (n = 72) in the three different displacement directions for the four subjects, six markers and three activities. It shows that the bras with different design features exhibited different levels of control. This gives some insight into the effective bra features.

Figure 6.

RBD of different bra styles in different directions.

It is evident from Figure 6 that Style 3 was, overall, the most effective in reducing breast displacement in all directions (62% in the x, 74% in the y and 76% in the z directions), followed by Style 2 (61% in the x, 60% in the y and 79% in the z directions). Referring to Table 4, the common design features of the effective bra samples of Styles 2 and 3 can be listed as follows:

compression type to limit breast movement in all directions;

short vest style with high neckline to maximize the coverage;

rigid cup seam to fit the breasts and prevent movement;

side slings to restrict the medial–lateral breast movement;

race-back panel to distribute the tension to the back;

slightly elastic bound neckline to fit the upper breast boundary for stabilization;

no centre gore, no cradle, no wire, no pad;

wide strap with good recovery, but with no adjustment.

The RBDs are also affected by the bra dimensions, as shown in Figure 7. In general, a higher centre front gave a larger anterior–posterior RBD in the x direction, wider shoulder straps provided a larger vertical RBD in the y direction and a higher neckline or deeper side seam also gave a larger medial–lateral RBD in the z direction.

Figure 7.

Relationships between bra measurement and RBD.

Based on Figure 7 and Table 5 the bra dimensions for size 75B that would give the most effective control of breast movement are suggested to be the following:

centre front height = 115 mm;

shoulder width = 30 mm;

neckline to bust point = 65 mm;

side depth = 65 mm;

underband width = 25 mm.

It is acknowledged that the different materials used in the construction of the bras will influence this. It is also noteworthy that the higher support level of sports bra claimed by the brand may not necessarily guarantee a more effective control of breast movement. In this study, Style 1 was labelled as a high level, while Style 2 was labelled as a medium level. However, the experimental results show that Style 2 provided a much higher RBD than Style 1, which is inconsistent with the manufacturer’s categorization of support levels.

The correlation coefficients for the elongation percentages of cup seam, neckline and shoulder straps (as mentioned in Table 4) and the RBD are shown in Table 8. Only the shoulder strap elongation percentage was significantly correlated with the RBD in the y direction (r = 0.82, n = 7).

Discussion

To answer research question Q1, the 3D breast movement relative to the thorax was measured in vivo for six marker positions on four woman subjects wearing seven different sports bras or no bra during three different activities (walking, jogging, and stepping up and down) using the validated BCS.

Comparing the displacements of the different breast markers, the movements of the inner and bottom breast markers were found to be significantly different from those of the nipple in all directions. Therefore, hypothesis H₁ was not rejected. The breast is a hyper-elastic body with various internal stresses. Any breast movement at one point will be influenced the others to a different extent. This showed that previous studies^{2,6,8,16,17,22,27,29} that used only one position to represent the whole breast movement were insufficient for research that aims to improve the design of different parts of sports bras. Therefore, in further scientific studies, investigations should also consider the detailed movement of different breast regions other than the nipple, especially in the inner and bottom breast quadrants. With a more comprehensive understanding of the 3D breast motion, delicate and integrative bra pattern designs can be recommended to provide the best fit and the most effective breast support in various activities. Measurement of only the nipple marker is insufficient for understanding how a bra functions.

Comparing the breast displacements during different activities, walking and running produced significantly different superior–inferior and medial–lateral displacements. However, there was no significant difference between stepping and running in all displacement directions. Therefore, H2 was only partly rejected. This also implies that the stepping exercise is considered to be not necessary in future studies if running is studied. In this study, the mean nipple displacement in sports bras was 8–11 mm during running at 7 km/h, smaller than the findings of Scurr et al.⁴³ of a mean of 14 mm in an encapsulation bra and 20 mm in a compression bra during the same speed of running. The difference is considered reasonable because smaller-breasted B-cup and C-cup Chinese women were participating in this study, while Scurr tested only D-cup subjects.

For research question Q2 (what are the effective features of sports bras in controlling breast movement), the key findings are stated and explained as follows.

The breast’s lateral movement is limited if a sling is used at the lateral side of each cup.

The control at the top part of the breast is most challenging. When the body moves upward, the shoulder strap loosens its tension and a gap between the bra and the breast can easily occur on the bra’s neckline. If the strap tension can be self-adjusted and kept in contact with the shoulders during the entire body movement, the tendency for such a gap to occur can be eliminated and the bra can better control the breast movement.

The most effective bra was Style 3. This was a firm-control short vest that had a unique inverted-U shape bounded seam over the upper and side boundaries of the cups. It strongly restricted the vertical and medial–lateral (z) breast movement, and held the breast tissues firmly around the chest, underarm and underbust with its wide cross-back design. The inelastic top front panel had a high neckline, so that the cups fully encaged the whole breasts. The cup panel used double bias cup seams to create the required volume and shape. A narrow-width fabric panel with a straight grainline passing through the nipple guaranteed the rigidity from the inner bottom breast to the outer top breast areas.

The least effective bra was Style 7. This had narrow shoulder straps (12 mm) and did not have a shoulder-to-back fabric panel, so it could not distribute the breast mass or transmit the tension effectively to the wearer’s shoulders and back. The rigid straps easily lost contact with the shoulders when the body moved upward, and thus the control became ineffective. The elastic seamless cups and low centre front allowed flexible movement. The low neckline could not hold the breasts in place and the upper parts of the breasts moved upward easily.

Regarding the research question Q3 (how can the design of sports bras be improved to reduce the breast movement), the key findings were the following.

Compression bras with a rigid and very high centre front (e.g. 115 mm in Styles 2 and 170 mm in Style 3) were the most effective in reducing the medial–lateral breast displacement. Both these firmly stabilized the two breasts against the chest wall, so as to reduce the distance from the centre of the breast mass to the chest wall, which, as a consequence, minimized the force moment and acceleration of breast movement, and thereby restricted the relative displacement between the breast tissues and the thorax muscles.

A large side depth (e.g. 100 mm in Style 6) effectively reduced the medial–lateral breast displacement, because the side breast tissues were kept fully within the inelastic cups, which prevented any lateral movement.

A front closure (Style 4) separated the two cups with a small connection at the centre front. Although this looked sexy, the bra cups did not firmly hold the breasts, so the medial–lateral breast movement was not well controlled.

Extensible shoulder straps are more effective in reducing the breast displacement, probably because they stay in contact with the body and keep elevating the breasts during the activities.

Summary and conclusion

The relative 3D breast displacement was analysed using a “Thorax–Breast Coordinate System”. The effectiveness of the sports bras was assessed by measuring their “RBD”. The results showed that there were significant differences in displacement between the nipple, inner breast and bottom breast, so using only a nipple marker is insufficient for a scientific analysis of detailed breast movement related to bra design. The bra samples were most effective in controlling the breast movement in the medial–lateral (z) direction, but were less effective in the anterior–posterior direction, so more attention should be paid to the design of the centre front, the plasticity of the cup fabric and the location of the cup seam and neckline. The poor control at the top part of the breast was due to a gap appearing between the breast and the inside of the top part of the cup, so it should be ensured that the neckline height is sufficient to fit the upper breast boundary.

The more effective bras shared common features, such as compression type, short vest style, high neckline, side slings, cross-back panel, bound neckline with a slight elasticity, no centre gore, no cradle, no wire, no pad and a non-adjustable wide strap. The most effective bra had a unique inverted-U shape bounded seam over the upper and side boundaries of the cups with a wide cross-back design. The inelastic top front panel was so high that the cups fully encaged the whole breasts. The cup panel used a narrow-width fabric panel with a straight grainline passing through the nipple to guarantee rigidity from the inner bottom breast to the outer top breast areas. Therefore, the breast movements in all directions were much reduced.

There is still no international standard for the evaluation of sports bras in reducing breast movement. To conclude, this paper has provided a basic insight into the requirements for such a standard and a guideline for consumers to select the most suitable running sports bra with appropriate features. The optimum bra measurements related to bra support were determined for the core size, which provided further insight into the design requirements for the optimum performance of sports bras. Even though the sample bra size of subjects was quite small, this preliminary study still provided a fairly comprehensive evaluation of the design features of sports bras, based on the scientific measurement of 3D breast displacement relative to the thorax during walking, running and stepping activities. Furthermore, the methodology will be useful for the future evaluation of other types of sportswear and functional garments. This original study gives a prelude to more systematic studies into the effects of individual components within the bra on its performance for a larger sample and wider range of breast sizes.

Footnotes

Acknowledgements

We would like to thank Dr Simon Harlock for his professional editing and comments.

Funding

We would like to thank the Research Grant Council for funding this research through the project PolyU 532306.

References

Hadi MSaA . Sports brassiere: Is it a solution for mastalgia? Breast J 2000; 6(6): 407–409.

Mason

Page

Fallon

. 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(2): 134–144.

McGhee

DEM

Steele

JRS

. Optimising breast support in female patients through correct bra fit: a cross-sectional study. J Sci Med Sport 2010; 13: 568–572.

Gehlsen

Albohm

. Evaluation of sports bras. Physician Sportsmed 1980; 8: 88–97.

Lorentzen

Lawson

. Selected sports bras: a biomechanical analysis of breast motion while jogging. Physician Sportsmed 1987; 15(5): 128–139.

Lawson

Lorentzen

. Selected sports bras: comparisons of comfort and support. Clothing and Text Research J 1990; 8(4): 55–60.

Himmelsbach

Valient

Lawson

. Peak breast accelerations when running different sports bras. Med Sci Sports Exer 1992; 24(Suppl): S187–S187.

Haycock

Shierman

Gillette

. The female athlete – does her anatomy pose problems? American Medical Association 19th conference on the medical aspects of sports, Monroe, Wisconsin, Illinois: AMA Press, 1978, pp. 1–8.

Okabe

Kurokawa

. Analysis of breast vibration while exercising and its reflection to vibration proof design of sports brassiere. Descente Sports Sci 2002; 23: 180–188.

10.

Okabe

Kurokawa

. Vibration characteristics of breasts during exercise with and without a brassiere. J Home Econ Jpn 2003; 54(9): 731–738.

11.

Okabe

Kurokawa

. Characteristics of three-dimensional displacement of the breasts wearing different kind of brassieres in young Japanese women. J Jpn Res Assoc Text End-Uses 2004; 45(6): 416–424.

12.

Okabe

Kurokawa

. Vibration and dislocation of the breast when wearing brassieres during running. J Home Econ pn 2005; 56(6): 379–388.

13.

Okabe

Kurokawa

. A study of the relationships between breast vibration, clothing pressure and dislocation under running conditions for designing sports brassiere. Descente Sports Sci 2006; 27: 75–85.

14.

Starr

Bransor

Shehab

. Biomechanical analysis of a prototype sports bra. J Text Apparel 2005; 4(3): 1–14.

15.

Fuseya

Matsumoto

. Three dimensional study on the vibration of breasts with brassieres. Wayo Joshi Daigaku Kiyo. Kaseikeihen 2006; 46: 1–12.

16.

Campbell

Munro

Wallace

. Can fabric sensors monitor breast motion? J Biomech 2007; 40(13): 3056–3059.

17.

Mcghee

Steele

Power

. Does deep water running reduce exercise-induced breast discomfort? Br J Sports Med 2007; 41(12): 879–883.

18.

White

Scurr

Smith

. The effect of breast support on kinetics during overground running performance. Ergonomics 2009; 52(4): 492–498.

19.

Mcghee

Steele

. Breast elevation and compression decrease exercise-induced breast discomfort. Med Sci Sports Exer 2010; 42(7): 1333–1338.

20.

Haake

Scurr

. A dynamic model of the breast during exercise. Sports Eng 2010; 12(4): 189–197.

21.

Scurr

White

Hedger

. The effect of breast support on the kinematics of the breast during the running gait cycle. J Sports Sci 2010; 28(10): 1103–1109.

22.

Gehlsen

Albohm

. Evaluation of sports bras. Physician Sportsmed 1980; 8(10): 88–97.

23.

Scurr

White

Hedger

. Three-dimensional displacement, velocity and acceleration of the breast during a running gait cycle. Med Sci Sports Exer 2008; 40(5): S221–S221.

24.

Scurr

White

Hedger

. Breast displacement in three dimensions during the walking and running gait cycles. J Appl Biomech 2009; 25: 322–329.

25.

Scurr J, Galbraith H, Hedger W, et al. A procedure for the analysis of three-dimensional breast displacement during treadmill exercise. 12th Annual Congress of the ECSS. Finland, 2007.

26.

Mason

Page

Fallon

. 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(2): 134–144.

27.

Eden

Valiant

Lawson

. Three dimensional kinematic evaluation of sport bra design. Med Sci Sports Exer 1992; 24(Suppl): S187–S187.

28.

Campbella

Munrob

Wallacea

. Can fabric sensors monitor breast motion? J Biomech 2007; 40(13): 3056–3059.

29.

Gehlsen

Stoner

. The female breast in sports and exercise. Med Sports Sci 1987; 24: 13–22.

30.

Shivitz NL. Adaptation of vertical ground reaction force due to changes in breast support in running, MSc Thesis, Oregon State University, 2002.

31.

Fan

Harlock

. Innovation and Technology of Women's Intimate Apparel, Boca Raton, Fla: CRC Press; Cambridge: Woodhead Publishing, 2006, pp. 171–172.

32.

Page

Steele

. Breast motion and sports brassiere design implications for future research. Sports Med 1999; 27(4): 205–211.

33.

Stamford

. Sports bras and briefs: choosing good athletic support. Physician Sportsmed 1996; 24(12): 99–100.

34.

Zhou

. Method of studying breast motion in sports bras: a review. Text Res J 2011; 81(12): 1234–1248.

35.

Zheng

. Breast sizing and development of 3D seamless bra. In: Institute of Textiles and Clothing, Hong Kong: The Hong Kong Polytechnic University, 2006, pp. 95–95.

36.

Selvik

. Roentgen stereophotogrammetry. A method for the study of the kinematics of the skeletal system. Acta Orthop Scand 1989; 60(232): 1–51.

37.

. The effects of menstrual cycle, site and individual variation on breast elasticity and thickness. Department of Health Technology and Informatics, Hong Kong: The Hong Kong Polytechnic University, 2009, pp. 90–95.

38.

Dixon

. Breast Ultrasound: How, Why and When, Edinburgh New York: Churchill Livingstone Elsevier, 2008.

39.

Ulger

Erdogan

Kumanlioglu

. Effect of age, breast size, menopausal and hormonal status on mammographic skin thickness. Skin Res Technol 2003; 9(3): 284–289.

40.

Snell

. Clinical Anatomy by Regions, Baltimore: Lippincott Williams and Wilkins, 2008.

41.

Zhou

. Studies of three-dimensional trajectories of breast movement for better bra design. Text Res J 2012; 82(3): 242–254.

42.

Zhou

. New methods to evaluate breast motion in braless and sports bra conditions. Institute of Textiles and Clothing, Hong Kong: The Hong Kong Polytechnic University, 2011.

43.

Scurr

White

Hedger

. Supported and unsupported breast displacement in three dimensions across treadmill activity levels. J Sports Sci 2011; 29(1): 55–61.

Bra style	Neckline to Bust point (mm)	Shoulder width (mm)	Underband width (mm)	Side depth (mm)	Centre front height (mm)
1	60	30	25	90	60
2	65	30	25	85	115
3	85	25	25	95	170
4	60	25	25	75	00
5	55	28	25	85	70
6	75	20	25	100	75
7	70	12	25	100	55

Bra style	Neckline to Bust point (mm)	Shoulder width (mm)	Underband width (mm)	Side depth (mm)	Centre front height (mm)
1	60	30	25	90	60
2	65	30	25	85	115
3	85	25	25	95	170
4	60	25	25	75	00
5	55	28	25	85	70
6	75	20	25	100	75
7	70	12	25	100	55

Bra style	Neckline to Bust point (mm)	Shoulder width (mm)	Underband width (mm)	Side depth (mm)	Centre front height (mm)
1	60	30	25	90	60
2	65	30	25	85	115
3	85	25	25	95	170
4	60	25	25	75	00
5	55	28	25	85	70
6	75	20	25	100	75
7	70	12	25	100	55