The perception of comfort influences orthosis therapy in a number of ways, such as in prescription and fitting (9) as well as in the adherence/compliance to ongoing use (2,10). It is influenced by many factors of which previous experience, presence of pain or injury, neurophysiology and psychological issues, and the environment are but a few (1,7,9).
Several authors have noted the perceived comfort of orthoses is strongly influenced by their design, but there is discordance as to which component of orthosis design determines comfort. A military study reported a significant decrease in the number of lower-limb injuries when recruits wore comfortable orthoses compared with their usual boots (8). Because the orthosis generally judged as most comfortable featured viscous material in the forefoot, the authors concluded that hardness of an orthosis is the dominant factor for comfort perception. This interpretation seemed to be supported by Chen et al. (1), who found that peak pressure, pressure-time integral, and contact area over the entire foot were significantly smaller for a comfortable cushioned, flexible orthoses compared with a harder uncomfortable design. Mundermann et al. (7) studied the relationship between the physiological effects of three orthoses of different posting and molding designs and their perceived comfort. They found that the orthosis that was molded without posting was associated with the largest changes in muscle activity and was also perceived to be the most comfortable. They also report that the effect of molding overrides the effect of posting and is thus the more important feature.
The use of a reliable scale is important when measuring comfort. Although the above studies used several different scales, a recent comparison of three commonly used comfort scales found a ranking scale to be the most reliable (6). Further, to obtain information regarding the amount of comfort, as opposed to relative comfort only, a 100-mm visual analog scale (VAS) was more reliable than a seven-point Likert scale. A change of 10.2 mm on the VAS was determined as representing clinically meaningful change. These scales have yet to be applied to orthoses.
The primary aim of this study was to investigate what features of an orthosis (i.e., contouring and hardness) and speed of gait (walking and jogging) influence perceived comfort. We hypothesized that: (a) individuals would find a flat orthosis more comfortable than contoured orthosis of identical density; (b) there would be an inverse relationship between hardness and comfort (i.e., as orthoses become softer, comfort would increase); and (c) there would be a change in perceived comfort with changing gait speeds.
Twenty subjects (8 females and 12 males, age = 26.8 ± 3.87 yr, weight = 72.67 ± 1.09 kg, height = 176.44 ± 8.42 cm) were recruited and consecutively allocated to two experimental protocols: one using the VAS and the other using the ranking scale. Subjects were recruited from local gymnasiums and excluded if they were injured at the time of the study, had received surgery, or had a history of neurological, sensory, or orthopedic conditions affecting the lower limbs. All subjects' shoes were inspected for wear (K.M.). If shoes showed excessive wear, i.e., torn upper or damage to the sole, subjects were advised to purchase and wear-in new shoes before they commenced the study, which occurred in one subject. Both protocols involved five sessions during five consecutive days. The study was approved by the University of Queensland medical research ethics committee and Australian Institute of Sport. All subjects were familiarized with the protocol before commencement, and written informed consent was obtained. All 20 subjects completed the study.
Four commercially available prefabricated orthoses (Vasyli International®, San Rafael, CA) were used in both experiments. Orthoses were full-length and constructed from ethylene-vinyl acetate (EVA) with fabric covering. They were fitted by a single physiotherapist (K.M.) and cut to fit the length of each subject's shoe. No further modifications took place. The three orthoses differed in hardness: hard (shore A 75°), medium (shore A 60°), and soft (shore A 52°) but exhibited the same contouring. A fourth orthosis was of uniform thickness along its length (3 mm) and was the same shore A value as the soft-contoured orthosis. Subjects were blinded to the differences between the orthoses.
Experiment 1: VAS
A horizontal 100-mm VAS, used in a previous study (6), was completed for the dimensions of overall comfort, forefoot cushioning, arch cushioning, and arch support. The terms "not comfortable at all" to "most comfortable imaginable" were used to anchor the scales.
After subjects had completed the VAS scale for each dimension, subjects were asked to judge each orthosis in relation to their shoe with the original sock-liner. For this criterion scale, the rating categories were "much less comfortable," "moderately less comfortable," "slightly less comfortable," "as comfortable as my shoe," "slightly more comfortable," "moderately more comfortable," and "much more comfortable."
Each session consisted of four trials. A trial involved subjects walking and jogging for 2 min on a treadmill, first in their usual jogging shoe then one of the four orthoses. The order of orthoses was randomized between subjects and between sessions. The speed of each trial was standardized between trials after being self-selected as a comfortable pace in the first session. As the orthoses varied in color depending on hardness, to ensure subjects remained blinded, the orthoses were kept out of the subjects' visual field while being inserted into their shoe.
Experiment 2: Ranking Scale
The ranking scale consisted of conditions 1 to 5 arranged vertically corresponding to the order the orthoses and shoe had been presented for that session. Subjects were asked to rank the five conditions (four orthoses and shoe) in order of overall comfort from 1 (most comfortable) to 5 (least comfortable). Subjects were permitted to make notes regarding each condition to assist in their choices within a session but were not permitted to view rankings or notes made from previous sessions.
Subjects traversed a 25-m runway, at self-selected comfortable speeds, in each orthosis and their shoe. The order of these conditions was randomized, and subjects were blinded to this order. Once all conditions were worn while walking or jogging, subjects were permitted to rewalk or jog in any of the conditions to be sure of their comfort rankings.
All analysis was conducted in SPSS version 16 (SPSS, Inc., Chicago, IL).
We have previously found that a minimum of two sessions is required to ensure reliable comfort measures on a VAS (6). Therefore, all data from the first session were excluded from our analysis.
Data from sessions 2 to 5 were entered into a mixed linear model with the main (fixed) effects of orthosis (hard, medium, and soft-contoured and soft flat) and gait (walking or jogging). Random variability between subjects and sessions was accounted for by including subject and trial as nested random effects. Significant main effects were followed up with tests of simple effects. Results are reported as mean difference with 95% confidence intervals (95% CI). Standardized mean difference (SMD = mean difference divided by the pooled SD) is also reported, referenced against Hopkins' (4) criteria of trivial (0.0-0.2), small (0.21-0.6), moderate (0.61-1.2), and large (>1.2).
In addition, all VAS scores of orthoses that corresponded to a judgment of slightly more, less, or as comfortable as my shoe were identified. A change score was calculated between these scores and the preceding score rated during wearing only the shoe. This was done because perceptions of comfort are often anchored in comparisons made to similar, previously experienced stimuli (8,11). Mean difference between change scores relating to slightly more or slightly less comfortable than my shoe and as comfortable as my shoe were compared with previously established minimal clinical important difference (MCID) (10.2 mm) (6) to examine the relationship between the VAS and criterion scale.
Data from sessions 1 to 5 were analyzed using nonparametric Friedman ANOVA (χ 2) to determine the effect of different orthoses. Post hoc Wilcoxon signed rank (T) was used to determine the difference between particular orthoses as well as between walking and jogging within a single orthosis design. Results of these comparisons are presented as the test statistic, which is the smallest of the two summed ranks, the significance level, and r denoting the effect size (z-value divided by the square root of the number of comparisons). This is also referenced to Hopkins' (4) criteria.
Experiment 1: VAS and Criterion-Based Rating Scale of Comfort
There are two parts to this experiment's results: first, there are the data from the rating of comfort on a VAS between the different shod conditions (i.e., shoe, flat orthoses, three orthoses of different hardness); and second, there is the relationship between the VAS in millimeters and a criterion scale anchored around the shoe-without-orthosis condition.
For all comfort dimensions, mixed modeling found significant main effects for the type of orthoses. The main effect of gait was also significant for overall comfort, arch cushioning, and arch support. The interaction between orthosis and gait was not significant (Table 1).
Pairwise comparisons found the soft-flat orthosis was statistically significantly more comfortable than all contoured orthoses including the soft orthosis on ratings of all comfort dimensions (Table 2). These differences were clinically meaningful (>10.2 mm) for overall comfort and arch cushioning. Mean differences between the hard and soft-flat orthoses (12.72 mm, 95% confidence interval = 6.42-19.01 mm) and the medium and soft-flat orthoses (14.99 mm, 95% confidence interval = 8.69-21.28 mm) were also clinically meaningful for ratings of arch support, as well as for forefoot cushioning (11 mm, 95% confidence interval = 7.98-14.04 mm and 11.19 mm, 95% confidence interval = 8.16-14.22 mm, respectively).
Statically significant differences were found between the medium and soft-contoured orthoses in overall comfort (7.19 mm, 95% confidence interval = 2.91-11.47 mm), arch cushioning (6.14 mm, 95% confidence interval = 0.54-11.75 mm), and arch support (7.39 mm, 95% confidence interval = 0.9-13.68 mm). There was also a difference between the overall comfort of the hard and soft orthoses (8.25 mm, 95% confidence interval = 3.97-12.53 mm). However, these were small effects (SMD < 0.33) and not clinically important.
Significant differences also occurred between walking and jogging for overall comfort (4.3, 95% confidence interval = 1.27-7.32), arch cushioning (5.24, 95% confidence interval = 1.28-9.21), and arch support (5.62, 95% confidence interval = 1.16-10.06) ratings of comfort (Table 3). These differences were not clinically important.
Relationship between the VAS and criterion scale.
From sessions 2 to 5, there were a total of 288 comparisons to the comfort of subjects' shoes for each comfort dimension (144 made in walking and 144 made in jogging). For measures of overall comfort and forefoot cushioning, most orthoses were considered to be ranging from slightly less comfortable than my shoe to slightly more comfortable than my shoe. For measures concerning the arch of the foot, subjects generally found the orthoses to be as comfortable as my shoe or less so (Table 4). Regardless of speed of gait, when subjects perceived the orthoses as slightly less comfortable than their shoe, the orthoses received a lower VAS score than the shoe. Likewise, when subjects perceived orthoses as slightly more comfortable, the VAS scores were, on average, higher than shoe ratings (Fig. 1). The pooled ratings of walking and jogging found that the difference between VAS scores corresponding to as comfortable as my shoes and slightly more comfortable was 11.34 mm. The mean difference between ratings of as comfortable as my shoe and slightly less comfortable was 17.49 mm.
Experiment 2: Relative Ranking of Comfort Between Orthoses
Friedman ANOVA indicated significant differences between the ranked positions of the orthoses (χ 2 3 = 21.98, P = 0.00). Post hoc analysis found the soft-flat orthosis was ranked higher and had greater perceived comfort than the hard (T = 1325, P < 0.001), the medium (T = 1297.5, P < 0.001), and soft-contoured orthoses (T = 1379.5, P < 0.001). The differences all had small effect sizes (r = 0.29, r = 0.3, and r = 0.27, respectively). There was no difference between the contoured orthoses (χ 2 2 = 0.15, P = 0.925).
The relative comfort of orthoses between walking and jogging was also investigated using Wilcoxon signed rank test. The ranked position of the orthoses did not change with different speed of gait: hard T = 184, P = 0.3, r = 0.1; medium T = 179.5, P = 0.255, r = 0.11; soft T = 176.5, P = 0.538, r = 0.06 and; flat T = 113, P = 0.928, r = 0.
The first aim of this study was to investigate what feature of an orthosis was the major consideration for comfort. Specifically, we hypothesized that a flat orthosis would be perceived as more comfortable than contoured orthoses. Second, that there would be an inverse relationship between hardness and comfort, such that the softer the orthosis the greater the perceived comfort. Third, that comfort of orthoses would change between walking and jogging. We also compared VAS comfort measures to criterion scale ratings to examine the relationship between the scales.
The soft-flat orthosis was found to be significantly more comfortable than all contoured orthoses using both the VAS and ranking scale. Although the effects of these differences are small based on the ranking scale, they are clinically important when compared with previously established MCID for footwear comfort using the VAS.
Between the contoured orthoses, we did not find any relationship between hardness and comfort using either scale. This is in contrast to the work of others (2,3) who reported a sequential increase in comfort with softer shoe midsoles or orthoses. The reason for this contrast may, among other things, be because of subjects being blinded to the difference between the orthoses in our study but not in other studies. Nevertheless, despite our finding to the contrary, there were small differences between the medium and soft orthoses on VAS dimensions of overall comfort, arch cushioning, and support and also between the hard and soft orthoses on ratings of overall comfort, but these were not clinically meaningful as adjudged by our criteria.
There were also small significant differences between VAS measures of walking and jogging for overall comfort, arch cushioning, and arch support. These were, however, not clinically important and not detected on the ranking scale. These results are in agreement with Chen et al. (1), who also found the relative comfort of four insoles did not change between walking and jogging. Clinically, the implication is that the relative comfort of the orthoses did not change between different speeds of gait and clinicians can be confident that a single orthosis is appropriate for both forms of gait.
In examining the relationship between the VAS and criterion scale, an increase in comfort of an orthosis beyond shoe comfort requires a change of 11.34 mm. However, to make more comfortable an orthosis that is perceived as less comfortable than a subject's usual footwear requires a change of 17.49 mm. Both of these amounts exceed the MCID of 10.2 mm, thus a clinically important change is required to change a rating on a criterion scale. It also indicates that the comfort VAS is not a linear scale, a similar finding to that of Wells et al. (12) in their study of rheumatoid arthritic patients. Clinically, it is harder to make an uncomfortable orthosis comfortable than it is to increase the comfort of an already comfortable orthosis. Wells et al. (12) suggest that changing orthosis comfort from uncomfortable to equal to a subject's shoe should be reported separately to a comfort increase in an already comfortable orthosis. By doing this, a more meaningful interpretation of results is possible.
There are limitations within this study that need to be acknowledged. First, the sample size is relatively small, so nonsignificant values may reflect type 2 error. However, we have referenced our results to a previously established MCID as a criterion for meaningful differences. As such, we believe our results reflect subjects' perceived changes in comfort beyond that which may be attributable to error. A second limitation is that we used a single brand of prefabricated orthoses. This was done to standardize material type (EVA), contouring of the orthoses, and fabrication. The material choice was based on previous evidence demonstrating EVA to be a comfortable choice (1) and a commonly used construct (5). Clinical application will, no doubt, benefit from further investigation into different types of contouring or degrees of posting. A third limitation is that our cohort was asymptomatic, which may limit generalizabilty to a symptomatic cohort. This was done to ensure that the focus of the study was the comfort of the foot-orthosis interface without the potential confounder of symptomatic relief. As a result, we were able to show clinically meaningful differences in comfort.
This study demonstrates that healthy subjects prioritize contouring over hardness when judging comfort because a soft flat or contoured orthosis was perceived to have greater overall comfort than medium and hard orthoses with identical contouring. Clinically meaningful changes are larger when needing to enhance comfort of uncomfortable orthoses than improve comfortable orthoses. Anchoring changes in comfort to the starting comfort level (i.e., uncomfortable or comfortable) is clinically important.
Financial support for this research was received from the Australian Research Council (Australian Research Council Linkage Project grant LP0668233). K.M. is supported by an Australian Research Council Australian Postgraduate Award Industry. Vasyli International® provided the inserts used in this study.
The authors report no conflicts of interest.
The results of this study do not constitute endorsement by the American College of Sports Medicine.
1. Chen H, Nigg BM, De Koning J. Relationship between plantar pressure distribution under the foot and insole comfort. Clin Biomech (Bristol, Avon)
2. Finestone A, Novack V, Farfel A, Berg A, Amir H, Milgrom C. A prospective study of the effect of foot orthoses composition and fabrication on comfort and the incidence of overuse injuries. Foot Ankle Int
3. Hennig EM, Valiant GA, Liu Q. Biomechanical variables and the perception of cushioning for running in various types of footwear. J Appl Biomech
5. Landorf K, Keenan AM, Rushworth RL. Foot orthosis
prescription habits of Australian and New Zealand podiatric physicians. J Am Podiatr Med Assoc
6. Mills K, Blanch P, Vicenzino B. Identifying clinically meaningful
tools for measuring comfort perception of footwear. Med Sci Sports Exerc
7. Mundermann A, Nigg BM, Humble RN, Stefanyshyn DJ. Orthotic comfort is related to kinematics, kinetics, and EMG in recreational runners. Med Sci Sports Exerc
8. Mundermann A, Nigg BM, Stefanyshyn DJ, Humble RN. Development of a reliable method to assess footwear comfort during running. Gait Posture
9. Mundermann A, Stefanyshyn DJ, Nigg BM. Relationship between footwear comfort of shoe inserts and anthropometric and sensory factors. Med Sci Sports Exerc
10. Nigg B, Nurse M, Stefanyshyn D. Shoe inserts and orthotics for sport and physical activities. Med Sci Sports Exerc
. 1999;31(7 suppl):S421-8.
11. Pawelka S, Kopf A, Zwick E, Bhm T, Kranzl A. Comparison of two insole materials using subjective parameters and pedobarography (PEDAR-System). Clin Biomech (Bristol, Avon)
12. Wells GA, Tugwell P, Kraag GR, Baker PR, Groh J, Redelmeier DA. Minimum important difference between patients with rheumatoid arthritis: the patient's perspective. J Rheumatol
Keywords:©2011The American College of Sports Medicine
ORTHOSIS; VISUAL ANALOG SCALE; RANKING SCALE; PERCEIVED COMFORT; CLINICALLY MEANINGFUL