An investigation into the influence of biological sex,anthropometrics,footwear,and dual tasking on balance

Abstract

BACKGROUND:

It remains unclear whether balance is influenced by biological sex, anthropometrics, wearing footwear, or dual tasking. Such information is important to aid clinical reasoning pertaining to assessment and rehabilitation.

OBJECTIVE:

To investigate the influence of biological sex, anthropometrics, footwear, physical activity and dual tasking (DT) on balance performance.

METHODS:

An observational study was performed on eighty-six healthy participants. Anthropometric assessment and static balance performance was evaluated, during double leg stance with eyes open (DLSEO) and eyes closed (DLSEC) and during single-leg-stance (SLS). All tasks were assessed with and without footwear and a cognitive task was introduced to assess the effect of DT on static balance performance.

RESULTS:

Generally, the static balance performance of females was better than males, across all balance tasks, with some large effect sizes (ES). In both sexes, without footwear tasks resulted in better balance during the DLSEC task but with footwear, static balance performance was better during SLS in males only. Overall minimal differences were observed between single and dual task with a large ES for SLS and DLSEO for females. Upper body size was moderately negatively correlated to static balance performance.

CONCLUSION:

Females outperformed males, footwear and DT had some, but minimal influence on static balance. Anthropometrics were moderately correlated with balance. Balance performance should be compared to unisex normative data sets and performed as a single task, with or without footwear.

Keywords

Correlation dual task footwear gender static balance

1 Introduction

Balance can be defined as the ability of an individual to maintain their centre of mass (COM) over their base of support [1]. It is regulated by the interaction of afferent inputs from the visual, vestibular and somatosensory systems, the central nervous system (CNS), the efferent output system and the musculoskeletal system [2, 3]. Balance is essential as a prerequisite for performing all activities of daily living and functional tasks and impairments in balance result in elevated injury risk [4] and loss of independence [5]. The assessment and training of balance is a fundamental component of many rehabilitation programmes [6], however, to optimise balance assessment and rehabilitation, several fundamental decisions need to be reasoned by thephysiotherapist.

Many intrinsic factors may influence balance, including those influencing the input systems (visual vestibular and somatosensory), the processing systems (brain, CNS) or the output system peripheral nervous system (PNS, muscles etc.) [7, 8]. Whether physiotherapists should compare balance performance with unisex groups remains a subject of debate. This is because some previous studies have suggested that biological sex influences balance, with studies demonstrating balance is moderated by biological sex [9], whilst others have shown this not to be the case [10]. Some discrepancies in the literature may be explained by anthropometric factors, rather than biological sex, where between sex differences were removed, once balance performance was normalised to height [11, 12]. However, no studies have systematically explored a range of anthropometric measurements and their influence on static balance performance. Such information is essential for the identifying balance impairment and establishing rehabilitation targets.

Whether or not footwear should be used in the assessment and rehabilitation of balance is a factor for consideration by physiotherapists and individuals. Footwear has the potential to affect static balance [13] through the modification of somatosensory input [14] or the physical constraint of foot and ankle movement [13]. Despite this, there is relatively little in the existing literature regarding the effect of footwear on balance, with some authors reporting footwear positively affects balance [15] and others reporting no effect [16]. These studies were limited to single sex cohorts and single leg stance (SLS) only and no studies have systematically explored the effect of footwear on balance across a range of tasks and conditions and such information is an imperative for clinical decision making.

The balance literature suggests that adding a cognitive-task results in significant additional balance challenge. However, the evidence of the effect of dual tasks (DT) on static balance is conflicted, with some suggesting better static balance [17], whilst others suggest no effect [18] or an adverse effect on balance [19]. The lack of clarity in the literature over the real influence of DT on static balance has led to confusion about the optimal approach to static balance assessment and training within research and clinical practice contexts. As the understanding of adding an additional task is not well understood, physiotherapists cannot be clear in their decision making regarding its use in impairment identification or rehabilitation/training.

Thus, this study aims to comprehensively investigate the influence of biological sex, anthropometrics, the wearing or not of footwear and the relative effects of single and DT on static balance performance during quiet standing. Such knowledge is critical for understanding how such factors influence balance, informing assessment and rehabilitation decisions.

This will be achieved by:

Exploring differences between females and males in static balance performance.

Investigating the relationship between anthropometrics and static balance performance.

Considering if wearing footwear affects static balance performance.

Determining if DT influences static balance performance.

This study will provide the foundation understanding to support decision making for static balance assessment and rehabilitation.

2 Methods

2.1 Study design

This study employs an observational design. A well-structured standardised protocol was applied to all participants and to minimise observer bias, the participants were kept unaware of the research aims during the trial.

2.2 Participants

A sample size calculation utilizing G power software was completed, based on an average effect size (ES) of 0.85 (data from Puszczalowska-Lizis et al. [20], with an alpha of 0.004 and a beta of 0.8, yielding the required sample size of 41 per group.

Participant groups of 42 males and forty-four females were enrolled, mean age 22.0±1.2 years, height 1.74±0.083 m, weight 78.3±0.023 kg and BMI (body mass index) 25.8±7.6 kg/m²; mean age 22.2±1.8 years, height 1.60±0.059 m, weight 62.5±0.012 kg, and BMI 24.4±4.2 kg/m², respectively. Recruitment was conducted via email, Twitter and advertisements at the Princess Nourah bint Abdulrahman, King Saud and Prince Sattam Bin Abdulaziz Universities, Saudi Arabia.

Healthy adults between the ages of 18 and 30 years old were recruited from university or college populations. Participants were excluded if they had experienced any lower limb injury during the previous 12 months, had current lower limb or lower back pain, a history of surgery, or rheumatological or neurological disorders that might affect balance, vestibular conditions, visual problems that may affect balance, were pregnant or unwell at the time of testing. All participants provided written informed consent prior to participating in accordance with the Declaration of Helsinki, approved by the Princess Nourah bint Abdulrahman University Institutional Human Ethics Committee. The data collection period ran from October 2020 to March 2021.

In addition to the demographic data, the participant’s anthropometrics were assessed during standing by an experienced physical therapist (over 10 years experience), including shoulder circumference (circumference inferior to the acromion process); waist circumference (circumference around the mid-point between the hip bones and the umbilicus) and hip circumference (circumference around the greater trochanters). The shoulder/waist ratio (SWR) was calculated, by dividing the shoulder circumference by the waist circumference, and the shoulder/hip ratio (SHR), by dividing the shoulder circumference by the hip circumference. Habitual levels of physical activity were also measured using the previously validated Baecke questionnaire [21].

2.3 Procedure

A rigorous checklist was applied, before, during and after the trial to ensure a standardised protocol. Every participant completed a 3-minute familiarisation with the tasks and the measurement device, the Prokin proprioceptive-stabilometric-assessment machine (Prokin 252, Technobody Inc, Italy). Static balance was then assessed during double leg stance (DLS) eyes open (DLSEO) and closed (DLSEC) and SLS eyes open (dominant leg determine by self-declaration), for 30 seconds each. Task order was randomised using a random number generator (www.random.org) to avoid the systematic learning effect.

Participants were instructed to balance, as best they could, on the Prokin platform with arms freely positioned for comfort. Foot position was standardised using a V-shaped marker, placed between the feet, located on the assigned marks on the Prokin platform. Sixty second rest periods were provided between each task. If the participant touched the safety bars or placed their non-balancing foot down or shifted their standing foot position, the task was considered a “fail” and was repeated; only two repeats were permitted. No verbal feedback was offered during the experiment. Static balance was assessed either with the participant wearing their own training shoes or without shoes (barefoot). Additionally, all the previous tasks were repeated with DT, where the participants were asked to count down from 100, subtracting 7, over 30 seconds. This mental calculation task is a common DT to be simultaneously executed during balance tasks [22].

2.4 Data capture and processing

Static balance was assessed using the Prokin 252 (Technobody Inc, Italy). Digitised data were obtained through integrated sensors and proprietary software. The Prokin device operated as a force plate, where the balance platform remained stationary and the centre of pressure (COP) was mapped across time at 20 Hz. The COP sway trace was used to generate both the ‘ellipse area’ of the sway trace and the ‘perimeter measurement’ of the sway trace. Ellipse area is the ellipse of best fit for 95% of the sway trace area (in mm²) and perimeter is the length of the sway trace (mm).

2.5 Statistical analysis

SPSS version 26 (SPSS, Version 26.0. Armonk, NY: IBM Corp) was used to analyse the data. All data were assessed for normality using the Shapiro-Wilk test. Since most of the data was non-normally distributed, Mann-Whitney U tests were used to explore differences between biological sex and Wilcoxon tests were used to determine differences between the with footwear and without footwear tasks and differences between the single and DT. Spearman’s correlations were used to investigate relationships between baseline characteristics and static balance performance. To determine if any differences across the biological sexes were due to height, balance scores were normalised for height and between sex comparisons were repeated.

Due to repeated pairwise comparisons, a Bonferroni correction was applied to reduce the alpha to 0.004 to compare between biological sex analysis and correlational analysis and 0.008, to compare between the with footwear and without footwear tasks and for single versus DT analyses. In addition, non-parametric effect size (ES) was calculated as outlined in Ricca and Blaine [23], with an ES≥0.9 considered large [24].

3 Results

3.1 Biological sex

There were statistically significant differences between females and males during static balance performances, with females consistently performing better. This was evident both with and without footwear, as well as single and DT, however, only the perimeter outcome measure reached statistical significance (Table 1). This was also the case after normalisation of balance scores with regard to height, for single leg stance and double leg stance with eyes closed without footwear, however, only the perimeter outcome measure reached statistical significance (Table 2).

Table 1
Results of static balance performance as measured by perimeter and ellipse area and statistical testing to determine if biological sex affects static balance performance with footwear (shod), without footwear (unshod) and during dual tasks (median and interquartile range)

Female perimeter (mm) Male perimeter (mm) Female ellipse area (mm²) Male ellipse area (mm²)

Task Med IQR Med IQR p-value ES Med IQR Med IQR p-value ES

Shod

DLSEOS 262 87 329 153 0.001* 0.94φ 121 115 155 204 0.056 0.53

DLSECS 367 147 480 204 <0.001* 1.03φ 212 207 311 283 0.040 0.75

SLSS 819 258 1019 331 <0.001* 1.15φ 401 217 558 384 0.005 0.91φ

Unshod

DLSEOUS 240 97 297 120 0.003* 0.91φ 110 137 121 199 0.171 0.18

DLSECUS 323 98 407 165 <0.001* 0.99φ 149 212 237 221 0.011 0.80

SLSUS 850 260 1122 433 <0.001* 1.09φ 450 357 546 271 0.037 0.55

Dual task

DLSEODTS 296 162 350 188 0.026 0.62 172 195 197 230 0.286 0.27

DLSECDTS 406 165 539 235 <0.001* 1.04φ 222 258 322 395 0.035 0.65

SLSDTS 863 384 1057 317 0.001* 1.08φ 536 301 612 256 0.051 0.54

DLSEODTUS 276 174 317 115 0.616 0.55 168 383 138 148 0.314 –0.39

DLSECDTUS 346 129 400 235 0.030 0.63 148 291 219 198 0.412 0.66

SLSDTUS 870 421 1083 298 0.001* 1.69φ 505 300 568 297 0.074 0.39

	Female perimeter (mm)	Male perimeter (mm)	Female ellipse area (mm²)	Male ellipse area (mm²)
Shod
DLSEOS	262	87	329	153	0.001*	0.94φ	121	115	155	204	0.056	0.53
DLSECS	367	147	480	204	<0.001*	1.03φ	212	207	311	283	0.040	0.75
SLSS	819	258	1019	331	<0.001*	1.15φ	401	217	558	384	0.005	0.91φ
Unshod
DLSEOUS	240	97	297	120	0.003*	0.91φ	110	137	121	199	0.171	0.18
DLSECUS	323	98	407	165	<0.001*	0.99φ	149	212	237	221	0.011	0.80
SLSUS	850	260	1122	433	<0.001*	1.09φ	450	357	546	271	0.037	0.55
Dual task
DLSEODTS	296	162	350	188	0.026	0.62	172	195	197	230	0.286	0.27
DLSECDTS	406	165	539	235	<0.001*	1.04φ	222	258	322	395	0.035	0.65
SLSDTS	863	384	1057	317	0.001*	1.08φ	536	301	612	256	0.051	0.54
DLSEODTUS	276	174	317	115	0.616	0.55	168	383	138	148	0.314	–0.39
DLSECDTUS	346	129	400	235	0.030	0.63	148	291	219	198	0.412	0.66
SLSDTUS	870	421	1083	298	0.001*	1.69φ	505	300	568	297	0.074	0.39

DLSECDTS; double leg stance eyes closed dual tasking shod, DLSECDTUS; double leg stance eyes closed dual tasking unshod, DLSECS; double leg stance eyes closed shod, DLSECUS; double leg stance eyes closed unshod, DLSEODTS; double leg stance eyes open dual tasking shod, DLSEODTUS; double leg stance eyes open dual tasking unshod, DLSEOS; double leg stance eyes open shod, DLSEOUS; double leg stance eyes open unshod, ES; effect size, IQR; interquartile range, Med; median, SLSDTS; single leg stance dual tasking shod, SLSDTUS; single leg stance dual tasking unshod, SLSS; single leg stance shod, SLSUS; single leg stance unshod, * significant at p≤0.004 level, φ Effect size ≥0.9.

Table 2

Results of static balance performance as measured by perimeter and ellipse area normalisation by height and statistical testing determining if biological sex affects static balance performance with footwear (shod), without footwear (unshod) conditions and during dual tasks (median and interquartile range)

	Female perimeter/height (mm/cm)		Male perimeter/height (mm/cm)		Female ellipse area/height (mm²/cm)		Male ellipse area/height (mm²/cm)
Task	Med/mean	IQR/SD	Med/mean	IQR/SD	p-value	ES	Med	IQR	Med	IQR	p-value	ES
Shod
DLSEOS	1.65	0.52	1.90	0.86	0.020	0.62	0.76	0.70	0.88	1.16	0.201	0. 32
DLSECS	2.33^a	0.63^b	2.76^a	0.74^b	0.005^c	0.07^d	1.36	1.18	1.74	1.57	0.132	0.56
SLSS	5.01	1.69	6.05	1.64	0.014	1.06φ	2.57	1.36	3.23	1.87	0.041	0.77
Unshod
DLSEOUS	1.50	0.66	1.70	0.78	0.068	0.52	0.67	0.86	0.70	1.10	0.439	0.10
DLSECUS	1.97	0.57	2.34	1.09	<0.001*	0.77	0.96	1.20	1.37	1.36	0.045	0.61
SLSUS	5.26	1.86	6.49	2.66	0.001*	0.88	2.82	2.34	3.12	1.69	0.220	0.30
Dual task
DLSEODTS	1.83	1.07	1.98	1.05	0.179	0.30	1.06	1.23	1.15	1.28	0.489	0.14
DLSECDTS	2.63^a	0.81^b	3.08^a	0.87^b	0.016^c	0.23^d	1.34	1.51	1.88	2.02	0.089	0.62
SLSDTS	5.39	2.33	6.02	1.93	0.075	0.60	3.27	1.86	3.53	1.52	0.303	0.29
DLSEODTUS	1.71	1.11	1.77	0.58	0.598	0.17	1.02	2.37	0.80	0.81	0.175	-0.48
DLSECDTUS	2.12	0.90	2.35	1.20	0.282	0.44	0.89	1.85	1.22	1.20	0.710	0.53
SLSDTUS	5.62	2.53	6.26	1.92	0.033	0.62	3.05	1.92	3.33	1.56	0.429	0.31

Large ES were observed for the differences between females and males for perimeter outcomes across all single balance tasks and some DTs, both with and without footwear (Table 1). Large ES were also observed for the differences between females and males, following normalisation for height for the perimeter outcomes across single leg stance with footwear (Table 2).

3.2 Anthropometry

There were significant moderate correlations between height, weight, shoulder and waist circumference with almost all DLS and SLS in both single and DTs (Table 3).

Table 3
Spearman’s rho (r) correlation between static balance performance and anthropometric characteristics

Task/parameter Height (r) Weight (r) Shoulder circumference (r) Waist circumference (r) Hip circumference (r) Shoulder-waist ratio (r) Shoulder-hip ratio (r) Physical activity (r)

Shod

DLSEOS perimeter .398* .317* .338* .392* .149 –.333* .159 .103

DLSEOS ellipse area .359* .394* .320* .370* .326* –.315* –.018 .131

DLSECS perimeter .436* .339* .368* .416* .114 –.374* .207 .110

DLSECS ellipse area .373* .341* .277 .299 .178 –.280 .023 .147

SLSS perimeter .286 .217 .294 .296 .061 –.256 .195 .038

SLSS ellipse area .412* .278 .295 .259 .082 –.169 .240 .185

Unshod

DLSEOUS perimeter .260 .338* .398* .380* .217 –.356* .053 .075

DLSEOUS ellipse area .285 .427* .320* .368* .358* –.338* –.037 .146

DLSECUS perimeter .487* .414* .477* .480* .190 –.361* .240 .132

DLSECUS ellipse area .385* .347* .328* .355* .194 –.285 .134 .117

SLSUS perimeter .344* .189 .309* .297 .019 –.237 .237 .031

SLSUS ellipse area .253 .222 .209 .260 .074 –.282 .108 –.024

Dual task

DLSEODTS perimeter .252 .135 .152 .201 –.044 –.212 .239 .012

DLSEODTS ellipse area .251 .218 .161 .184 .096 –.179 .113 –.041

DLSEODTUS perimeter .135 .122 .051 .118 .062 –.183 .040 .022

DLSEODTUS ellipse area .064 .132 .002 .060 .142 –.132 –.089 .075

DLSECDTS perimeter .380* .331* .363* .404* .124 –.341* .187 .073

DLSECDTS ellipse area .320* .350* .287 .357* .177 –.348* .078 .088

DLSECDTUS perimeter .246 .117 .128 .191 –.056 –.191 .209 .122

DLSECDTUS ellipse area .195 .215 .107 .192 .083 –.213 .044 .189

SLSDTS perimeter .325* .185 .228 .243 .046 –.201 .152 –.066

SLSDTS ellipse area .269 .156 .148 .136 .030 –.104 .155 .036

SLSDTUS perimeter .370* .272 .322* .340* .137 –.257 .188 –.072

SLSDTUS ellipse area .313* .381* .308* .377* .241 –.330* .077 .011

Task/parameter	Height (r)	Weight (r)	Shoulder circumference (r)	Waist circumference (r)	Hip circumference (r)	Shoulder-waist ratio (r)	Shoulder-hip ratio (r)	Physical activity (r)
Shod
DLSEOS perimeter	.398*	.317*	.338*	.392*	.149	–.333*	.159	.103
DLSEOS ellipse area	.359*	.394*	.320*	.370*	.326*	–.315*	–.018	.131
DLSECS perimeter	.436*	.339*	.368*	.416*	.114	–.374*	.207	.110
DLSECS ellipse area	.373*	.341*	.277	.299	.178	–.280	.023	.147
SLSS perimeter	.286	.217	.294	.296	.061	–.256	.195	.038
SLSS ellipse area	.412*	.278	.295	.259	.082	–.169	.240	.185
Unshod
DLSEOUS perimeter	.260	.338*	.398*	.380*	.217	–.356*	.053	.075
DLSEOUS ellipse area	.285	.427*	.320*	.368*	.358*	–.338*	–.037	.146
DLSECUS perimeter	.487*	.414*	.477*	.480*	.190	–.361*	.240	.132
DLSECUS ellipse area	.385*	.347*	.328*	.355*	.194	–.285	.134	.117
SLSUS perimeter	.344*	.189	.309*	.297	.019	–.237	.237	.031
SLSUS ellipse area	.253	.222	.209	.260	.074	–.282	.108	–.024
Dual task
DLSEODTS perimeter	.252	.135	.152	.201	–.044	–.212	.239	.012
DLSEODTS ellipse area	.251	.218	.161	.184	.096	–.179	.113	–.041
DLSEODTUS perimeter	.135	.122	.051	.118	.062	–.183	.040	.022
DLSEODTUS ellipse area	.064	.132	.002	.060	.142	–.132	–.089	.075
DLSECDTS perimeter	.380*	.331*	.363*	.404*	.124	–.341*	.187	.073
DLSECDTS ellipse area	.320*	.350*	.287	.357*	.177	–.348*	.078	.088
DLSECDTUS perimeter	.246	.117	.128	.191	–.056	–.191	.209	.122
DLSECDTUS ellipse area	.195	.215	.107	.192	.083	–.213	.044	.189
SLSDTS perimeter	.325*	.185	.228	.243	.046	–.201	.152	–.066
SLSDTS ellipse area	.269	.156	.148	.136	.030	–.104	.155	.036
SLSDTUS perimeter	.370*	.272	.322*	.340*	.137	–.257	.188	–.072
SLSDTUS ellipse area	.313*	.381*	.308*	.377*	.241	–.330*	.077	.011

DLSECDTS; double leg stance eyes closed dual tasking shod, DLSECDTUS; double leg stance eyes closed dual tasking unshod, DLSECS; double leg stance eyes closed shod, DLSECUS; double leg stance eyes closed unshod, DLSEODTS; double leg stance eyes open dual tasking shod, DLSEODTUS; double leg stance eyes open dual tasking unshod, DLSEOS; double leg stance eyes open shod, DLSEOUS; double leg stance eyes open unshod, r; spearman’s rho correlation, SLSDTS; single leg stance dual tasking shod, SLSDTUS; single leg stance dual tasking unshod, SLSS; single leg stance shod, SLSUS; single leg stance unshod, * significant at p≤0.004 level.

Shoulder/waist ratio was moderately correlated to almost all DLS balance variables for both the with footwear condition, the without footwear condition and single tasks. During DTs there was a significant moderate correlation between the shoulder/waist ratio in the DLSEC with footwear and SLS without footwear conditions (Table 3).

All directions of effect regarding these correlations suggest greater height, weight and shoulder and waist circumferences are related to poorer static balance (Table 3). There were no significant correlations between total physical activity level and the static balance variables (Table 3).

3.3 Footwear versus without footwear

Across the majority of comparisons, footwear made little difference. However, differences were seen for DLSEC, where performance was better without footwear in females and males, as measured using the sway perimeter (Table 4). Performance with footwear was better than without footwear, yielding a significant difference during SLS for males only (Table 4). Despite these differences, no large ES were determined.

Table 4
Results of significance testing and effect size calculation for the effect of being with footwear (shod) or without footwear (unshod) on static balance performance

Gender/parameter DLSEO shod vs unshod DLSEC shod vs unshod SLS shod vs unshod

p-value ES p-value ES p-value ES

Female

Perimeter (mm) 0.091 –0.44 <0.001* –0.70 0.890 0.25

Ellipse area (mm²) 0.111 –0.25 0.009 –0.75 0.433 0.39

Male

Perimeter (mm) 0.019 –0.56 0.006* –0.77 0.004* 0.58

Ellipse area (mm²) 0.016 –0.63 0.007* –0.70 0.57 –0.08

Gender/parameter	DLSEO shod vs unshod	DLSEC shod vs unshod	SLS shod vs unshod
Female
Perimeter (mm)	0.091	–0.44	<0.001*	–0.70	0.890	0.25
Ellipse area (mm²)	0.111	–0.25	0.009	–0.75	0.433	0.39
Male
Perimeter (mm)	0.019	–0.56	0.006*	–0.77	0.004*	0.58
Ellipse area (mm²)	0.016	–0.63	0.007*	–0.70	0.57	–0.08

DLSEC; double leg stance eyes closed, DLSEO; double leg stance eyes open, ES; effect size, SLS; single leg stance, * significant at p≤0.008 level.

3.4 Single versus DT

Significant differences emerged for the DLSEC with footwear condition in both genders and without footwear in females only; static balance performance was better during the single task, as measured by the sway perimeter (Table 5). Additionally, for DLSEO without footwear in females only, performance was better during the single task (Table 5).

Table 5
Results of significance testing and effect size calculation for the effect of single and dual tasking on static balance performance

Gender/parameter Shod single task vs dual task Unshod single task vs dual task

DLSEO DLSEC SLS DLSEO DLSEC SLS

p-value ES p-value ES p-value ES p-value ES p-value ES p-value ES

Female

Perimeter (mm) 0.016 –0.60 0.003* –0.47 0.360 –0.32 <0.001* –0.73 0.001* –0.41 0.427 –0.12

Ellipse area (mm²) 0.024 –0.78 0.340 –0.10 0.019 –1.23φ 0.002* –0.92φ 0.116 0.01 0.068 –0.41

Male

Perimeter (mm) 0.075 –0.29 0.004* –0.53 0.423 –0.25 0.684 –0.38 0.159 0.08 0.247 0.22

Ellipse area (mm²) 0.129 –0.55 0.925 –0.08 0.179 –0.34 0.375 –0.30 0.764 0.17 0.406 –0.16

Gender/parameter	Shod single task vs dual task	Unshod single task vs dual task
Female
Perimeter (mm)	0.016	–0.60	0.003*	–0.47	0.360	–0.32	<0.001*	–0.73	0.001*	–0.41	0.427	–0.12
Ellipse area (mm²)	0.024	–0.78	0.340	–0.10	0.019	–1.23φ	0.002*	–0.92φ	0.116	0.01	0.068	–0.41
Male
Perimeter (mm)	0.075	–0.29	0.004*	–0.53	0.423	–0.25	0.684	–0.38	0.159	0.08	0.247	0.22
Ellipse area (mm²)	0.129	–0.55	0.925	–0.08	0.179	–0.34	0.375	–0.30	0.764	0.17	0.406	–0.16

DLSEC; double leg stance eyes closed, DLSEO; double leg stance eyes open, ES; effect size, SLS; single leg stance, * significant at p≤0.008 level, φ; Effect size≥0.9.

A large ES was observed for SLS with footwear (Ellipse area) and DLSEO without footwear (Ellipse area) in females only. However, no large ES were observed in males (Table 5).

4 Discussion

This study aimed to clarify the influences of biological sex, anthropometrics, physical activity, footwear and DT on static balance performance during quiet standing, to provide evidential support for decision making in the assessment and rehabilitation of static balance.

Females outperformed males across all static balance performance tests, with some very large ES. Previous studies mirror those of the current study, with women performing better than men during static balance [12 , 26]. Thus, the current study not only compliments the previous literature but demonstrates that this difference is present regardless of whether shoes are worn and whether single or DTs are performed during testing.

The reasons for the sex related differences during static balance performance are unknown. It is possible that anthropometric factors play a role, as it has been suggested that differences in balance between the sexes is merely an expression of the height of the individual [11 , 27], since being taller results in a higher COM, however, differences were also found after normalisation for height. The current study established modest correlations with height but also weight in healthy adults during static balance during quiet standing. This correlation may help to explain the results, since males tend to have a relatively larger upper body, whilst females have greater mass concentrated in the lower body [28, 29]. However, additionally, waist and shoulder circumferential measures showed a modest effect on static balance performance, suggesting the greater the ‘size’ of the upper body including the waist, the poorer the balance performance across DLS tasks.

This larger upper body size is likely to raise the COM and potentially contributes to greater sway. Our correlational analysis seems to support a relationship between greater upper body size (albeit moderately) and poorer balance performance. This was particularly the case with DLS tasks. It is possible that such anthropometric relationships are evident during simpler balance tasks, as almost universally, SLS tasks were not significantly correlated with body size. Previous research has demonstrated that DLS moderately correlated with BMI [11] and the components of BMI [30], suggesting the significance of anthropometrics is evident in lower complexity tasks.

The current study showed no correlation between total physical activity level and any static balance variables. This lack of relationship suggests that modifications to physical activity level will have little direct effect of balance ability. This is in agreement with previous authors, who did not find any correlation between sports performance level and static balance in athletes of any age [31].

To the best of the authors’ knowledge the current study is the first to investigate the effect of footwear on static balance performance in healthy adults under multiple positions and conditions. This information is essential for physiotherapists when deciding whether to assess and train static balance with or without footwear. To determine the optimal approach, it is first required to determine if balance performance differs across these conditions. The findings here demonstrate that overall, there were no differences in performance during static balance due to footwear. Two exceptions were found: (i) during DLSEC tasks wearing shoes worsened performance in both females and males; and (ii) wearing shoes resulted in better balance performance for males under the SLS condition. Pressure moving across the distribution of the plantar foot surface during static balance appears to be translated into movement around the COM, which could impact balance performance. Previous studies have demonstrated that plantar foot surface sensation is a critical contributor to the somatosensory part of the biofeedback loop involved in maintaining balance [32]. Arguably, this may be enhanced during the barefoot condition, as supported by findings in this study for the DLS condition but not the SLS condition. The difference may be explained by visual contribution. Potentially, the DLSEC condition may have resulted in a re-evaluation of the weighting of the balance input parameters to the biofeedback loop [33]. With vision present (SLS condition), reliance on plantar pressure feedback is unnecessary/less important, due to the dominance of the visual contribution. Without vision (DLSEC condition) it is possible that attention to the input from the plantar foot pressure is raised, resulting in the difference between both with and without footwear conditions. The current literature is surprisingly sparse regarding direct comparisons between the with and without footwear conditions; therefore, the current study offers a significant contribution to the current understanding. Previous studies have demonstrated mixed results regarding the effect of footwear on balance. Germano et al. [15] demonstrated similar findings for SLS, reporting COP excursion was greater in the barefoot condition. In contrast, Smith et al. [34] demonstrated that balance, whilst wearing shoes, was better for COP sway overall and in the anteroposterior direction during DLSEC. The reason for these conflicting findings is not clear; however, angular displacement of centre of gravity (COG) was used to determine balance performance, rather than perimeter length of the sway trace [34].

Significant differences were determined for DLSEC with footwear and without footwear conditions, demonstrating poorer static balance performance during DT. DT balance testing has increasingly been reported in the literature, since it affords greater discriminatory capacity than single task balance tests [35]. However, in this study, while performance was marginally worse with DT, no significant differences were determined except during the DLSEC task. This likely reflects the participants’ age, as the current study’s young participants have adequate capacity to complete a cognitive task with little ‘cost’ to the physical task. Previous literature, demonstrating a difference, focuses on older adults who have relatively less capacity to manage the competing demands of attending to both the cognitive task and the balance task [17]. Previous studies found no significant differences between single and DT during static balance performance, as measured by total COP displacement, ellipse area and COP velocity in the force plate in youths [18]. Therefore, it seems unlikely that the DT offers additional benefit over single task balance testing for young healthy individuals.

This study has made several contributions to understanding balance assessment and performance through the systematic exploration of biological sex, footwear, DT and correlational analysis, linking performance to baseline anthropometrics. This information is beneficial for physiotherapists when determining how best to test balance and prescribe balance exercises. Balance assessments should be completed with an understanding of female-male differences and comparing female or male specific reference scores is recommended. Wearing footwear seems to negatively affect females and males for DLSEC and males positively for SLS; however, the ES was small, demonstrating a minimal difference in actual values between the conditions. This suggests use of footwear has minimal clinical influence on balance performance. DT for routine balance testing in young healthy adults offers little, beyond single task balance testing in terms of the detection of balance deficits. In addition, this study offers insights that explain balance performance from which physiotherapists can integrate into their exercise justification. Clinicians might advise individuals to increase their physical activity but this is likely to have a minimal effect on balance performance, because of the lack of a relationship between physical activity and balance performance. A person’s height, weight, waist and shoulder circumferences did relate to balance performance; therefore, an awareness of the increased burden anthropometric characteristics may place on balance should be acknowledged.

4.1 Limitations

The participants in this study were all young, healthy, and unimpaired, therefore, care must be taken when applying these findings to other populations. The balance exercises were repeated; therefore, it is possible that fatigue or a learning effect had some impact on the outcomes, however, this was likely minimised due to task order randomisation.

5 Conclusion

In conclusion, females demonstrated better static balance across a range of tasks. The sex differences noted were present in both with and without footwear conditions, as well as in single and DTs. No large ES were observed between with footwear and without footwear conditions or between single and DTs for static balance. Regarding anthropometry, correlations were moderate between static balance performance and height, weight and upper body size. It is concluded that the taller, heavier and greater the size of the upper body, the poorer static balance performance will be. Such information can be utilised by physiotherapists to facilitate decision making when assessing performance and prescribing rehabilitation exercise.

Footnotes

Acknowledgments

The first author gratefully acknowledges the Princess Nourah bint Abdulrahman University for funding her PhD program at Cardiff University. Also thanks to Drs Ali Albarati, Abdulrahman Alsubiheen, Mazyad Alotaibi, Afrah Almuwais, Khaled AlGhamdi, Mrs Zahra Al Asiri and Mr.Abdulellah Alkhanfour for facilitating recruitment for this study.

Conflict of interest

The authors have no conflict of interest to report.

Ethical approval

According to the Declaration of Helsinki, approved via the Princess Nourah bint Abdulrahman University Institutional Human Ethics Committee, with this number H-01-R-059.

Funding

The first author was funded by the Saudi Cultural Bureau in United Kingdom, London and Princess Nourah bint Abdulrahman University in Saudi Arabia, Riyadh.

References

Fusco

, Giancotti

, Fuchs

, Wagner

, Varalda

, Capranica

, et al. Dynamic balance evaluation: reliability and validity of a computerized wobble board. J Strength Cond Res. 2020;34(6):1709–15. doi: 10.1519/JSC.0000000000002518

Horak

. Clinical measurement of postural control in adults. Phys Ther. 1987;67(12):1881–5. doi: 10.1093/ptj/67.12.1881

Shumway

, Horak

. Assessing the influence of sensory interaction on balance. Suggestion from the field. Phys Ther. 1986;66(10):1548–50. doi: 10.1093/ptj/66.10.1548

Emery

, Cassidy

, Klassen

, Rosychuk

, Rowe

. Effectiveness of a home-based balance-training program in reducing sports-related injuries among healthy adolescents: a cluster randomized controlled trial. CMAJ. 2005;172(6):749–54. doi: 10.1503/cmaj.1040805

Dunsky

. The effect of balance and coordination exercises on quality of life in older adults: a mini-review. Front Aging Neurosci. 2019;11:318. doi: 10.3389/fnagi.2019.00318

Thompson

, Arena

, Riebe

, Pescatello

. ACSM’s new preparticipation health screening recommendations from ACSM’s guidelines for exercise testing and prescription 9th edition. Curr Sports Med Rep. 2013;12(4):215–7. doi: 10.1249/JSR.0b013e31829a68cf

Campbell

, Borrie

, Spears

. Risk factors for falls in a community-based prospective study of people of 70 years and older. J Gerontol. 1989;44(4):M112–7. doi: 10.1093/geronj/44.4.m112

Tinetti

. Performance-oriented assessment of mobility problems in elderly patients. J Am Geriatr Soc. 1986;34(2):119–26. doi: 10.1111/j.1532-5415.1986.tb05480.x

Era

, Sainio

, Koskinen

, Haavisto

, Vaara

, Aromaa

. Postural balance in a random sample of 7,979 subjects aged 30 years and over. J Gerontol. 2006;52(4):204–13. doi: 10.1159/000093652

10.

Palazzo

, Nardi

, Lamouchideli

, Caronti

, Alashram

, Padua

, et al. The effect of age, sex and a firm-textured surface on postural control. Exp Brain Res. 2021;239(7):2181–91. doi: 10.1007/s00221-021-06063-2

11.

, Abu Osman

, Yusof

, Wan Abas

WAB

. Biomechanical evaluation of the relationship between postural control and body mass index. J Biomech. 2012;45(9):1638–42. doi: 10.1016/j.jbiomech.2012.03.029

12.

Maki

, Holliday

, Fernie

. Aging and postural control: a comparison of spontaneous- and induced-sway balance tests. J Am Geriatr Soc. 1990;38(1):1–9. doi: 10.1111/j.1532-5415.1990.tb01588.x

13.

Runge

, Rehfeld

, Resnicek

. Balance training and exercise in geriatric patients. J Musculoskelet Neuronal Interact. 2000;1(1):61–5. Available from: https://pubmed.ncbi.nlm.nih.gov/15758528/

14.

Lee

, Cho

, Lee

. Short-foot exercise promotes quantitative somatosensory function in ankle instability: a randomized controlled trial. Med Sci Monit. 2019;25:618–26. doi: 10.12659/MSM.912785

15.

Germano

AMC

, Schlee

, Milani

. Balance control and muscle activity in various unstable shoes compared to barefoot during one-leg standing. Footwear Sci. 2012;4(2):145–51. doi: 10.1080/19424280.2012.674063

16.

Alghadir

, Zafar

, Anwer

. Effect of footwear on standing balance in healthy young adult males. J Musculoskelet Neuronal Interact. 2018;18(1):71–5. Available from: https://pubmed.ncbi.nlm.nih.gov/29504581/

17.

Boisgontier

, Beets

IAM

, Duysens

, Nieuwboer

, Krampe

, Swinnen

. Age-related differences in attentional cost associated with postural dual tasks: increased recruitment of generic cognitive resources in older adults. Neurosci Biobehav Rev. 2013;37(8):1824–37. doi: 10.1016/j.neubiorev.2013.07.014

18.

Lüder

, Kiss

, Granacher

. Single- and dual-task balance training are equally effective in youth. Front Psychol. 2018;9(6):912. doi: 10.3389/fpsyg.2018.00912

19.

Yardley

, Gardner

, Leadbetter

, Lavie

. Effect of articulatory and mental tasks on postural control. Neuroreport. 1999;10(2):215–9. doi: 10.1097/00001756-199902050-00003

20.

Puszczalowska-Lizis

, Bujas

, Jandzis

, Omorczyk

, Zak

. Inter-gender differences of balance indicators in persons 60–90 years of age. Clin Interv Aging. 2018;13:903–12. doi: 10.2147/CIA.S157182

21.

Baecke

JAH

, Burema

, Frijters

JER

. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr. 1982;36(5):936–42. doi: 10.1093/ajcn/36.5.936

22.

Paillard

, Noé

. Techniques and methods for testing the postural function in healthy and pathological subjects. Biomed Res Int. 2015;2015:891390. doi: 10.1155/2015/891390

23.

Ricca

, Blaine

. Notes on a nonparametric estimate of effect size. J Exp Educ. 2020;90(1):249–58. doi: 10.1080/00220973.2020.1781752

24.

Senington

, Lee

, Williams

. Biomechanical risk factors of lower back pain in cricket fast bowlers using inertial measurement units: a prospective and retrospective investigation. BMJ Open Sport Exerc Med. 2020;6(1):e000818. doi: 10.1136/bmjsem-2020-000818

25.

Ekdahl

, Jarnlo

, Andersson

. Standing balance in healthy subjects. Evaluation of a quantitative test battery on a force platform. Scand J Rehabil Med. 1989;21(4):187–95. Available from: https://pubmed.ncbi.nlm.nih.gov/2631193/

26.

Mickle

, Munro

, Steele

. Gender and age affect balance performance in primary school-aged children. J Sci Med Sport. 2011;14(3):243–8. doi: 10.1016/j.jsams.2010.11.002

27.

Bryant

, Trew

, Bruce

, Kuisma

RME

, Smith

. Gender differences in balance performance at the time of retirement. Clin Biomech. 2005;20(3):330–5. doi: 10.1016/j.clinbiomech.2004.11.006

28.

Farenc

, Rougier

, Berger

. The influence of gender and body characteristics on upright stance. Ann Hum Biol. 2003;30(3):279–94. doi: 10.1080/0301446031000068842

29.

Menegoni

, Galli

, Tacchini

, Vismara

, Cavigioli

, Capodaglio

. Gender-specific effect of obesity on balance. J Obes. 2009;17(10):1951–6. doi: 10.1038/oby.2009.82

30.

Chiari

, Rocchi

, Cappello

. Stabilometric parameters are affected by anthropometry and foot placement. Clin Biomech. 2002;17(9-10):666–77. doi: 10.1016/s0268-0033(02)00107-9

31.

Andreeva

, Melnikov

, Skvortsov

, Akhmerova

, Vavaev

, Golov

, et al. Postural stability in athletes: the role of age, sex, performance level, and athlete shoe features. Sports. 2020;8(6):89. doi: 10.3390/sports8060089

32.

McKeon

, Hertel

, Bramble

, Davis

. The foot core system: a new paradigm for understanding intrinsic foot muscle function. Br J Sports Med. 2015;49(5):290–9. doi: 10.1136/bjsports-2013-092690

33.

Benjuya

, Melzer

, Kaplanski

. Aging-induced shifts from a reliance on sensory input to muscle cocontraction during balanced standing. J Gerontol A Biol Sci Med Sci. 2004;59(2):166–71. doi: 10.1093/gerona/59.2.M166

34.

Smith

, Burton

, Johnson

, Kendrick

, Meyer

, Yuan

. Effects of wearing athletic shoes, five-toed shoes, and standing barefoot on balance performance in young adults. Int J Sports Phys Ther. 2015;10(1):69–74. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/pmc4325290/

35.

Ruffieux

, Keller

, Lauber

, Taube

. Changes in standing and walking performance under dual-task conditions across the lifespan. Sports Med. 2015;45(12):1739–58. doi: 10.1007/s40279-015-0369-9