Bruening et al.

BMC Musculoskeletal Disorders (2019) 20:608

https://doi.org/10.1186/s12891-019-2981-6

TECHNICAL ADVANCE Open Access

Functional assessments of foot strength: a

comparative and repeatability study
Dustin A. Bruening1* , Sarah T. Ridge1, Julia L. Jacobs1, Mark T. Olsen1, Dallin W. Griffin1, Drew H. Ferguson1,
Kirk E. Bassett2 and A. Wayne Johnson1

Abstract
Background: Evaluating the strength of the small muscles of the foot may be useful in a variety of clinical
applications but is challenging from a methodology standpoint. Previous efforts have focused primarily on the
functional movement of toe flexion, but clear methodology guidelines are lacking. A novel foot doming test has
also been proposed, but not fully evaluated. The purposes of the present study were to assess the repeatability and
comparability of several functional foot strength assessment techniques.
Methods: Forty healthy volunteers were evaluated across two testing days, with a two-week doming motion
practice period between them. Seven different measurements were taken using a custom toe flexion dynamometer
(seated), custom doming dynamometer (standing), and a pressure mat (standing). Measurements from the doming
dynamometer were evaluated for reliability (ICCs) and a learning effect (paired t-tests), while measurements from
the toe flexion dynamometer and pressure mat were evaluated for reliability and comparability (correlations).
Electromyography was also used to descriptively assess the extent of muscle isolation in all measurements.
Results: Doming showed excellent within-session reliability (ICCs > 0.944), but a clear learning effect was present,
with strength (p < 0.001) and muscle activity increasing between sessions. Both intrinsic and extrinsic muscles were
engaged during this test. All toe flexion tests also showed excellent reliability (ICCs > 0.945). Seated toe flexion tests
using the dynamometer were moderately correlated to standing toe flexion tests on a pressure mat (r > 0.54);
however, there were some differences in muscle activity. The former may better isolate the toe flexors, while the
latter appeared to be more functional for many pathologies. On the pressure mat, reciprocal motion appeared to
display slightly greater forces and reliability than isolated toe flexion.
Conclusions: This study further refines potential methodology for foot strength testing. These devices and
protocols can be duplicated in the clinic to evaluate and monitor rehabilitation progress in clinical populations
associated with foot muscle weakness.
Keywords: Doming, Short foot exercise, Toe flexion, Intrinsic foot muscles, Extrinsic foot muscles

Background Recent research has shown that foot muscle weakness

The evaluation of foot muscle strength may be import- may be associated with increased fall risk in the elderly
ant in a variety of clinical diagnostic and rehabilitation due to the muscles’ role in stabilization and balance [11,
applications [1]. For instance, diseases such as diabetes 12]. Emerging evidence also suggests that targeted foot
[2] and Charcot-Marie-Tooth [3] contribute to foot strengthening exercises may benefit those with foot
weakness and altered foot neuromuscular activation [4]. pathologies [13, 14]. The ability to accurately assess foot
Foot muscle weakness is also associated with a wide strength may help clinicians and researchers identify
range of foot deformities and foot pain (e.g. [5–10]). weakness and monitor rehabilitation progress.
While robust methods for evaluating the strength of
muscle groups crossing the major body joints is com-
* Correspondence: dabruening@byu.edu
1
Exercise Sciences Department, Brigham Young University, Provo, UT, USA
monplace, evaluating the strength of the small foot mus-
Full list of author information is available at the end of the article cles is challenging. In particular, the intrinsic foot

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 2 of 9

muscles contribute to many of the same actions as larger Methods

extrinsic lower limb muscles, making it difficult to isolate Participants
their effects. Previous efforts have focused primarily on Forty healthy participants (20 M, 20 F) were recruited
the functional movement of toe flexion, which targets a from Brigham Young University’s campus. Sample size
number of the smaller foot muscles (although still a com- was calculated using Gpower software (version 3.1.9.2)
bination of intrinsic and extrinsic muscles). Toe flexor and based on estimates of differences in doming as well
strength methodology has relied on hand held dynamom- as correlations between toe flexion tests. The former was
etry [9, 15, 16] or more qualitative tests such as the Paper estimated in part from a previous study [22], doubling
Grip Test or Intrinsic Positive Test [1, 17–20]. These the inter-day doming differences to account for training.
methods contain substantial subjective components and This yielded an effect size of 0.5; with 80% power and
may thus be limited in reliability. Mickle et al. [12, 21] α = 0.05, a sample size of 34 was estimated. Similarly, de-
suggested exerting downward force on a pressure mat, tecting a correlation effect size of r = 0.5 with the same
creating a quantitative measurement system which could power yielded a sample size estimate of 29. 40 partici-
also separate the hallux from the rest of the phalanges pants were chosen to provide additional power.
during the same functional movement. Members of our To ensure that all participants were healthy and free from
research group recently presented a custom built toe foot pain, participants were asked if they had any ortho-
flexion device wherein participants grip a small bar with pedic injuries or impairments that might affect foot func-
their toes and pull against a force transducer [22]. These tion and strength, with affirmative responses excluded. All
latter two options both appear promising from a reliability participants were volunteers and signed consent forms
standpoint, but could elicit different muscular actions. A approved by the Brigham Young University institutional re-
functionality comparison between them may help clarify view board in compliance with the Declaration of Helsinki.
their utility in strength assessments. Sample demographics were: height = 172.2 ± 9.8 cm,
In our previous paper [22], we also described a novel weight = 72.3 ± 13.1 kg, age = 25.3 ± 6.2 yrs.
method for assessing the strength of a doming, or short-
foot [23], exercise. This movement involves contracting Strength tests
the quadratus plantae and other muscles crossing the Five strength testing protocols were performed on three dif-
midfoot to raise the medial longitudinal arch (MLA) and ferent apparatus. For analysis, this resulted in a total of
thus “shorten” the foot. This movement has been used seven assessed strength measurements: 1) Great Toe
as a rehabilitation tool with a goal of engaging and Flexion (GTF), 2) Lateral Toe Flexion (LTF), 3) Great Toe
strengthening the intrinsic foot muscles [24], but accom- Pressure Isolated (GTPI), 4) Lateral Toe Pressure Isolated
panying strength assessments have not been published. (LTPI), 5) Great Toe Pressure Reciprocal (GTPR), 6) Lateral
Our doming device used a force transducer connected Toe Pressure Reciprocal (LTPR), and 7) Doming (DOM).
to a padded cuff placed over the dorsal foot to measure Both GTF and LTF used the same device, which was de-
upwards MLA force. We tested within session reliability, scribed in detail previously [22]. Briefly, participants sat in
which was quite good [22]. However, while toe flexion is a chair with their knees at a 90° angle and testing foot flat
somewhat habitual for most adults, doming can be an on a board that spanned from the heel to 1st metatarsal
awkward movement that may require training to achieve head. Participants then gripped a carabiner with the great
a reliable measurement across days. The possibility of a toe (GTF) or a T-bar with the 2nd and 3rd toes (LTF) and
learning effect has not been assessed. pulled on a cable attached to a force transducer (model
The overall purpose of the present study was to test SMT1–22, Interface, inc. Scottsdale AR, USA) (Fig. 1a).
the repeatability and comparability of the aforemen- One modification was made to the previous design - the
tioned methods for assessing foot muscle strength. For transducer was raised slightly to reduce the toe flexion
toe flexion, our main goal was comparative, to determine angle required to hold the carabiner T-bar. A turnbuckle
whether toe flexion on a pressure mat and toe flexion on attached between the transducer and carabiner was ad-
our custom-built device targeted similar muscle func- justed until a baseline tension between 1.5–3.0 N was
tions. For doming, our main goal was to assess reliability achieved in a stable, but not exerted, position. Upon a sig-
across days and evaluate the extent of a potential learn- nal from the investigator, participants pulled with max-
ing effect. For both movements, we also sought to pre- imal effort sustained over a three second period.
liminarily assess the extent to which participants were GTPI, LTPI, GTPR, and LTPR were all performed on a
isolating the targeted muscles through the use of surface pressure mat (HR Mat, Tekscan inc., Boston MA, USA)
electromyography (EMG). Robust, reliable, and targeted in a manner patterned after Mickel et al. [21]. GTPI and
methods of assessing foot strength may open new re- LTPI came from the same tests, as did GTPR and LTPR.
search opportunities to understand foot function and For the former, participants stood in the center of the
new clinical opportunities to affect this function. pressure mat and were asked to isolate and press the
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 3 of 9

Fig. 1 Three strength testing apparatus were used to create seven strength assessments. a) Toe flexion device (GTF and LTF), b) Pressure mat
(GTPI, LTPI, GTPR, and LTPR), c) Doming (DOM) – the force transducer is in tension underneath the wooden platform

hallux and toes of the testing foot as hard as they could the cuff to the transducer inside the box and were tight-
into the pressure mat, holding for three seconds. Total ened using a turnbuckle to a pre-load of 4.5 N - 5.5 N.
force under each foot was monitored during the test, Doming was accomplished by shortening the foot and
with no more than 60% of body weight allowed on the pushing the arch upwards against resistance, exerting
testing foot. The reciprocal tests were identical to the maximally for three seconds. Instructions were given to
isolated tests, but the contralateral toes were extended slide the ball of the foot towards the heel without curling
concurrently with flexion of the testing toes (Fig. 1b). the toes. Participants were given several practice trials
For both tests, force under the great toe and lateral toes prior to collection. Note that following study completion,
was separated during analysis. the molded plastic cuff was replaced with a leather cuff,
The device used for DOM was modified more exten- and the turnbuckle was replaced with a Boa enclosure sys-
sively from that previously described [22]. The transducer tem (Boa Technology inc., Denver, CO, USA), which im-
(Model LC101–100, Omegadyne, inc., Stamford CT, proved comfort and ease of use (Fig. 1c).
USA) was moved from above the foot to below it, creating
a pulling action with the transducer in tension instead of Protocol
compressive. This was done to alleviate any possibility of Data collection was performed on two separate days,
misalignment between foot motion and force transducer two weeks apart. The strength testing protocol for both
axis. Participants stood on a small wooden box, placing days was identical, with one exception. On the first day,
the testing foot in a brannock device. A molded plastic a second session was performed on the DOM test, to
cuff was then placed over the dorsal foot. Wires secured better assess acute learning effects. Height, weight and
foot dominance were also documented on the first day.
Testing was performed on the dominant foot only,
which was determined by foot preference in 2 out of 3
activities (step-up, ball kick, and response to push).
Prior to strength testing, five wireless, bipolar, Ag/AgCl
surface electrodes (Trigno, Delsys Inc., Boston, MA, USA)
were affixed to the skin surface of the dominant foot,
using double-sided tape. Locations corresponded to the
muscle bellies of the medial gastrocnemius (GS), fibularis
longus (FL), fibularis brevis (FB), tibialis anterior (TA),
and abductor hallucis (AH) (Fig. 2). Two functional move-
ments were then performed to elicit near-maximal muscle
contractions and establish reference values for EMG acti-
vation. The first was a single-leg calf raise (with handheld
Fig. 2 EMG electrode placement. Surface electrodes were placed assist), held for three seconds. This movement was used
over the tibialis anterior (TA), fibularis longus (FL), fibularis brevis (FB),
as a reference for the GS, FL, FB, and AH muscles. In the
abductor hallucis (AH), and medial gastrocnemius (GS)
second movement, participants balanced on both heels,
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 4 of 9

dorsiflexing their ankles and raising their toes off the shape, the operator moved the window to the most suit-
ground, again holding for three seconds. This was used as able location, visually balancing high force with low vari-
a reference for the TA muscle. EMG was collected in all ability. The extracted one-second average force metric
tests at 1926 Hz. was used for statistical analysis.
Three trials of each of the five strength tests were per- Data from the pressure mat was initially processed in
formed barefoot on the dominant foot. The order of the Tekscan’s FootMat software (v. 7.1). Masks were manu-
five tests was randomized using a practical two-tiered ally drawn around the great toe and lateral toes, and
approach (i.e. split-plot design). First, the order of the total force under each of these areas, expressed as a per-
three apparatus were randomized (doming device, toe centage of total force, was exported. This was then
flexion device, and pressure mat); next the order of the multiplied by body mass to convert to force units. Re-
individual tests performed at each device were random- gion forces were then imported into the same LabView
ized (e.g. GTF and LTF). For each trial, participants were software and analyzed in the same manner as the trans-
instructed to exert maximal effort and hold for three ducer forces.
seconds. Data from the devices using force transducers EMG signals were also processed in LabView. All sig-
(GTF, LTF, and DOM) were collected at 1000 Hz using nals were zero-offset, low pass filtered (350 Hz 4th order
LabView (National Instruments DAQ 9171, Austin TX Butterworth), rectified, and high pass filtered (5 Hz 4th
USA), while data from the pressure mat was collected at order Butterworth) to get a linear envelope. EMG signals
100 Hz through the manufacturer’s software interface. were then time-synchronized with their associated force
After the first day, participants were given instructions profiles (Fig. 3b), and the average envelope amplitudes
to practice the doming motion prior to the second day. during the same one-second windows were extracted.
A minimum of 10 exercise bouts were required during For the force transducers, the electrodes were collected
the two weeks between days, each bout consisting of 2 through the same software, allowing for direct software
sets of 10 contractions. Investigators contacted the par- synchronization. For the pressure mat, however, EMG
ticipants on a daily basis to provide reminders. While and pressure were collected separately, initiated by a ver-
the second test day was instituted primarily to assess po- bal command. Synchronization was fine-tuned after-
tential learning effects for the DOM tests, all tests were wards by manually aligning the onset of muscle
repeated for consistency, and the protocol for day two activation with the increase in force production. For
was therefore identical to day one (with the exception of presentation, EMG amplitudes were normalized to their
the second DOM session). associated reference near-maximal contractions and
expressed as percentages.
Data processing
Data from the force transducers was analyzed in custom Statistical analysis
LabView software (National Instruments, Austin TX Repeatability of all tests was assessed using intraclass cor-
USA). Transducer force was plotted over time with a relation coefficients (ICC) with random participants and
movable one-second window designated on the graph. fixed raters for absolute agreement as well as Standard
The program automatically placed the window over the Error of the Measurement (SEm). These were calculated
region with the highest one-second average force across the three trials for each session independently.
(Fig. 3a). However, if this did not exhibit a plateau-like Mean differences between each day were compared using

Fig. 3 Example doming force (a) and EMG (b) plots. Force and EMG were extracted over a 1-s time window (shaded area). The onset and
termination of force are marked with dotted lines, which were aligned with the same events in the EMG signal for synchronization. Mean EMG
magnitude was calculated over the same time window
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 5 of 9

paired t-tests, except for DOM, which contained three Mean EMG activation levels across tasks varied by
total sessions. For this, a one-way repeated measures muscle (Fig. 5). GS EMG activity was small during all
ANOVA was used, followed by LSD post-hoc pairwise tests, but largest in the pressure mat tests (~ 17% max).
comparisons. Comparability of toe flexion strength be- TA activity ranged from 12% (GTPI / LTPI) to 45%
tween the toe flexion device and the pressure mat tests (DOM) max. FL and FB had comparable activation levels
was done using Pearson correlation coefficients. Signifi- in DOM (~ 36% max), while FB activity was slightly
cance for all tests was set at α = 0.05. Descriptive statistics greater than FL across all toe flexion tasks (~ 26% vs ~
were used to evaluate muscle activation during each test. 16% max). AH activity ranged from 25% (LTF) up to
For DOM, EMG data from all three sessions are pre- 94% (DOM) max. In DOM, all muscle activations ap-
sented, while for the other tests only the data from the peared to increase across the three sets.
second day tests are presented.
Discussion
The purposes of this study were to assess repeatability
Results and comparability of several functional foot strength as-
All devices showed excellent within-session repeatability sessment techniques. For discussion, doming is sepa-
(Table 1), with ICC values between 0.945 (GTPI, day 1) rated from the various toe flexion tests.
and 0.986 (LTPR, day 2). ICCs for all tests increased
slightly between day 1 and day 2. Toe flexion
While the mean force values (Table 2) for all tests also Both the toe flexion device and the pressure mat appear
increased slightly between day 1 and day 2, only in to be reliable tools for assessing toe flexor muscle
DOM did this reach a statistically significant difference. strength, albeit with some functional differences between
Pairwise comparisons among the three DOM sets them. Within session repeatability was high for both de-
showed a significant change between day 2 and both sets vices and both were moderately correlated to each other.
of day 1 (p < 0.001), but no difference between the two Notably, the correlation plots were more dispersed at
sets on day 1 (p = 0.111). higher forces (Fig. 4), suggesting that the execution strat-
Both toe flexion tests were significantly correlated with egies diverged between devices as force increased. Also,
their respective pressure mat tests (p < 0.001) (Fig. 4). the forces under the hallux were higher than the lateral
GTF was moderately correlated with both GTPI (r = toes on the toe flexion device, while the reverse was true
0.54) and GTPR (r = 0.55), while LTF was slightly more, for the pressure mat. Correlation coefficients between
but still moderately, correlated with both LTPI (r = 0.65) devices were also higher for the lateral toes than the hal-
and LTPR (r = 0.64). lux. It may be that the instruction to flex all toes at one
time on the pressure mat resulted in a strategy that un-
equally focused on the lateral toes. EMG also suggests
Table 1 Measurement repeatability
that foot musculature was being activated to a slightly
Measure Day 1 Day 2
different degree between devices. In particular, the pres-
DOM S1: 0.944 S2: 0.974 0.973
sure mat engaged the AH more. This is consistent with
(0.903–0.969) (0.953–0.986) (0.956–0.984) the greater lateral toe pressure, and is due in part to the
GTF 0.961 0.984 standing posture [23]. Because of the need for
(0.931–0.979) (0.974–0.991) stabilization and balance when standing, the pressure
LTF 0.965 0.982 mat may represent a more functional movement, while
the toe flexion device likely better isolates the toe flexor
(0.942–0.980) (0.969–0.990)
muscles. The latter may be more preferable when asses-
GTPI 0.945 0.974
sing specific changes in muscle strength, due to fewer
(0.913–0.967) (0.958–0.984) compensatory strategies. However, participants with
LTPI 0.949 0.962 compromised distal motor control, deformities, or sensa-
(0.918–0.969) (0.939–0.977) tion loss (e.g. diabetic neuropathy) may find it easier to
GTPR 0.952 0.983 press downward on the pressure mat than to grip the
toe flexion device.
(0.923–0.971) (0.972–0.990)
When using a pressure mat to assess toe flexion, a
LTPR 0.952 0.986
number of factors suggest that reciprocal motion may be
(0.923–0.971) (0.978–0.992) preferable over isolated motion. The reciprocal test had
Intraclass Correlation Coefficients are shown along with associated 95% slightly higher ICCs as well as higher mean forces. It also
Confidence Intervals in parentheses. All measures contained one set of three
trials on both days, with the exception of DOM, which included two sets on
appeared to be subjectively more comfortable, as the re-
day 1 (S1 and S2) ciprocal motion helped the participant maintain a stable
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 6 of 9

Table 2 Raw force data (mean ± standard deviation) for each measure, expressed in Newtons. P-values for paired t-tests between
days are shown. For DOM, the p-value is from a full one-way ANOVA across all three sets. Pairwise comparisons are displayed below
the table
Measure (N) Day 1 Day 2 P-value
DOM S1: 117.9 ± 72.4 S2: 131.0 ± 72.0 164.2 ± 86.0 < 0.001 (ANOVA)*
GTF 50.0 ± 22.0 53.1 ± 27.0 0.193
LTF 41.9 ± 18.5 46.8 ± 24.3 0.517
GTPI 51.6 ± 32.6 56.0 ± 35.7 0.137
LTPI 86.4 ± 41.3 93.4 ± 45.5 0.121
GTPR 57.1 ± 32.4 62.6 ± 43.9 0.142
LTPR 95.7 ± 42.5 102.3 ± 57.0 0.231
* Pairwise comparisons: p = 0.111 for day 1, set 1 vs. 2; p = 0.001 for day 1 set 2 vs day 2; p < 0.001 for day 1 set 1 vs day 2

posture with even weight distribution between feet, other studies reported substantially higher values than
avoiding a forward lean. This is not surprising given the ours, ranging from 60 to 130 N [16, 27], likely due to dif-
heavy use of reciprocal motion in the lower extremity ferent positioning (e.g. standing or lying supine), partici-
during walking, running, and other functional movements pant samples, and equipment. We chose our testing
– the nervous system appears to be wired for reciprocal methodology in a specific effort to isolate toe flexor
motion in the lower extremity, in contrast with more mir- muscles, and the lower force values may represent suc-
rored movements in the upper extremity [25, 26]. cessful isolation. The previous toe flexion dynamometry
Our toe flexion force magnitudes differ from previ- studies [16, 27] are actually closer in magnitude to our
ously reported values, most likely due to methodological pressure mat values, which may be in part due to the
differences. In our own previous study [22], we reported standing posture. Our pressure protocol was based on
slightly higher toe flexion forces than the current study, Mickle et al. [13]; however, we showed lower pressures
perhaps due to the subtle design modifications. Most under the hallux (18% BW compared to 8% BW) and

Fig. 4 Correlations between the toe flexion device and pressure mat. Left column = isolated tests, right column = reciprocal tests; top row = great
toe, bottom row = lateral toe: a) GTPI vs GTF, b) GTPR vs GTF, c) LTPI vs LTF, d) LTPR vs LTF. See Methods for specific acronym names
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 7 of 9

Fig. 5 Mean muscle EMG activation levels. Comparisons among all toe flexion tests, day 2 (left), and among the three sessions of doming (right)

higher pressures under the lateral toes (11% BW com- accustomed to the motion is not clear from this study, but
pared to 14% BW). The main difference between the it is clear that some amount of practice should be
studies was that Mickle used separate tests for the hallux employed for clinical utility.
and lateral toes. We chose to capture force using a single While the GS was mostly inactive during the doming
test in an attempt to reduce the involvement of compen- motion, the TA was clearly used by the participants, and
satory or stabilizing muscles that may be engaged when its activation also appeared to increase with familiarity.
trying to isolate the great toe or lateral toes. For ex- Both the TA and the tibialis posterior (TP) cross the
ample, hallux extension or hindfoot inversion could in- midfoot and are extrinsic facilitators of MLA rise. While
crease lateral toe force by unloading the medial side of it is difficult to fully isolate the AH from the TA and TP,
the foot. Both ours and Mickle’s values are slightly the AH is clearly active during the doming movement,
higher than Menz et al. [19], who employed the paper particularly when it is performed standing [23]. Doming
grip test over a similar pressure mat. has been proposed as a rehabilitation exercise to
strengthen intrinsic foot muscles, in part because it en-
Doming gages the AH more than toe flexion does (ref 23 here
Due to its novelty, there is little previous research with too). However, the extrinsic TA and TP muscles can
which to compare our doming strength data. In contrast overpower or overcompensate for weak intrinsic muscles
to our toe flexion tests, our doming values are slightly [28], causing an existing muscle imbalance to persist
higher than our previous study [22], but likely for similar even after doming training. Future studies should focus
reasons (modifications made to the device interface). on training strategies that isolate the AH from the TA
While within session repeatability was excellent in both and TP.
studies, DOM showed a clear learning effect, as forces and
ICCs both increased between day 1 and day 2. EMG re- Limitations and future directions
veals some insight into this learning. With the exception The main study limitations revolved around the use of
of the GS, the activity of the other muscles mirrored the surface EMG to assess muscle isolation. This yielded
force, increasing with each successive set and motion fa- some insights, but could not fully assess intrinsic muscle
miliarity. The AH in particular showed a large increase, activity; this would require fine-wire EMG. In addition,
even between the two sets on day 1. The AH is the pri- EMG activity was normalized to a functional maximum.
mary intrinsic foot muscle responsible for MLA rise, and This was chosen because individual muscle maximums
it is not surprising that increased familiarity with the awk- are difficult to obtain for the feet. Jung et al. [23, 29]
ward doming movement increased its role as well as used a manual muscle test to represent maximum AH
DOM strength values. While the FL and FB do not dir- contraction – this adds tester variability and may be an
ectly contribute to MLA rise, their activity also increased, unfamiliar movement for many participants, but should
likely due to the increased need for stabilization to control be considered in future work. We also did not fully as-
foot inversion that accompanies MLA rise. Accustomiza- sess the repeatability of the strength tests between ses-
tion appeared to happen even between sessions within a sions or between raters; however, this was performed
day – as evidenced by the combination of higher (al- previously [22]. We also chose not to normalize the
though not significant) mean forces, ICCs, and muscle ac- force values to give readers a sense of the strength mag-
tivity. The amount of practice required to become fully nitudes and normalization was not needed for our
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 8 of 9

within-subject comparisons. However, appropriate nor- Author details

1
malizations (e.g. to body weight) may need to be investi- Exercise Sciences Department, Brigham Young University, Provo, UT, USA.
2
Mechanical Engineering Department, Brigham Young University, Provo, UT,
gated, for example when comparing strength changes USA.
among participants (i.e. between-within designs). An-
other future direction is to make use of the variability in Received: 19 February 2019 Accepted: 29 November 2019

the force-time waveform to investigate force steadiness

(e.g. [30, 31]), which may provide insight into motor
References
dysfunction. 1. Soysa A, Hiller C, Refshauge K, Burns J. Importance and challenges of
measuring intrinsic foot muscle strength. J Foot Ankle Res. 2012;5(1):29.
Conclusions 2. Andersen H. Motor dysfunction in diabetes. Diabetes Metab Res Rev. 2012;
28:89–92.
Overall, this study further established several tools that 3. Gallardo E, García A, Combarros O, Berciano J. Charcot–Marie–Tooth disease
can be used to quantify foot muscle strength. For toe type 1A duplication: spectrum of clinical and magnetic resonance imaging
flexion strength evaluation, our recommendation for features in leg and foot muscles. Brain. 2005;129(2):426–37.
4. Fernando M, Crowther R, Lazzarini P, Sangla K, Cunningham M, Buttner P,
choosing between dynamometry and pressure mat tests Golledge J. Biomechanical characteristics of peripheral diabetic neuropathy:
depends on the clinical application. The toe flexion de- a systematic review and meta-analysis of findings from the gait cycle,
vice is likely better for isolation and assessing changes muscle activity and dynamic barefoot plantar pressure. Clin Biomech. 2013;
28(8):831–45.
over time, while the pressure test is more functional and 5. Allen RH, Gross MT. Toe flexors strength and passive extension range of
easier for some populations to perform. For the latter, motion of the first metatarsophalangeal joint in individuals with plantar
we recommend using a reciprocal motion between fasciitis. J Orthop Sports Phys Ther. 2003;33(8):468–78.
6. Bus SA, Maas M, Michels RP, Levi M. Role of intrinsic muscle atrophy in the
limbs. Practically, the toe flexion device can be con- etiology of claw toe deformity in diabetic neuropathy may not be as
structed inexpensively, but requires a dedicated appar- straightforward as widely believed. Diabetes Care. 2009;32(6):1063–7.
atus and software. Pressure mats are often used clinically 7. Cheung R, Sze L, Mok N, Ng G. Intrinsic foot muscle volume in experienced
runners with and without chronic plantar fasciitis. J Sci Med Sport. 2016;
for gait and posture assessments, however, foot strength 19(9):713–5.
analysis requires some custom post-processing. For 8. Incel NA, Genc H, Erdem H, Yorgancioglu Z. Muscle imbalance in
doming, our novel test could be duplicated in the clinic, hallux valgus: an electromyographic study. Am J Phys Med Rehabil.
2003;82(5):345–9.
but would also require a separate apparatus and soft- 9. Kwon O, Tuttle L, Johnson J, Mueller M. Muscle imbalance and reduced
ware. We are currently working on an integrated design ankle joint motion in people with hammer toe deformity. Clin Biomech.
that combines toe flexion and doming in a single device. 2009;24(8):670–5.
10. Saeki J, Nakamura M, Nakao S, Fujita K, Yanase K, Ichihashi N. Muscle
We recommend that instruction and practice of the stiffness of posterior lower leg in runners with a history of medial tibial
doming motion be provided to patients prior to initial stress syndrome. Scand J Med Sci Sports. 2018;28(1):246–51.
testing. 11. Kelly LA, Kuitunen S, Racinais S, Cresswell AG. Recruitment of the plantar
intrinsic foot muscles with increasing postural demand. Clin Biomech. 2012;
27(1):46–51.
Acknowledgements
12. Mickle KJ, Munro BJ, Lord SR, Menz HB, Steele JR. Toe weakness and
We would like to thank the following individuals for their assistance with
deformity increase the risk of falls in older people. Clin Biomech. 2009;
data collection and processing: Adrienne Henderson, Kennedy Poplawski,
24(10):787–91.
Matt McClung, Austin Rasmussen, Joseph Droge, Ben Keith, and Jonathan
13. Mickle KJ, Caputi P, Potter JM, Steele JR. Efficacy of a progressive resistance
Egbert.
exercise program to increase toe flexor strength in older people. Clin
Biomech. 2016;40:14–9.
Authors’ contributions 14. Jung D-Y, Koh E-K, Kwon O-Y. Effect of foot orthoses and short-foot exercise
DAB, STR, and AWJ planned the study, analyzed the data, and wrote the on the cross-sectional area of the abductor hallucis muscle in subjects with
manuscript. JLJ and MTO contributed to study design, data collection, data pes planus: a randomized controlled trial 1. J Back Musculoskelet Rehabil.
analysis, and manuscript writing. DWG, DHF, KEB contributed to study design
2011;24(4):225–31.
and collected data. All authors revised and approved the final manuscript. 15. Senda M, Takahara Y, Yagata Y, Yamamoto K, Nagashima H, Tukiyama H,
Inoue H. Measurement of the muscle power of the toes in female
Funding marathon runners using a toe dynamometer. Acta Med Okayama. 1999;
There was no external funding associated with this study. 53(4):189–91.
16. Spink MJ, Fotoohabadi MR, Menz HB. Foot and ankle strength assessment
Availability of data and materials using hand-held dynamometry: reliability and age-related differences.
The data generated by this study will be available from the corresponding Gerontology. 2010;56(6):525–32.
author on reasonable request. 17. de Win MM, Theuvenet WJ, Roche PW, de Bie RA, van Mameren H. The
paper grip test for screening on intrinsic muscle paralysis in the foot of
Ethics approval and consent to participate leprosy patients. Int J Lepr Other Mycobact Dis. 2002;70(1):16–24.
The study was approved by the institutional review board at Brigham Young 18. Garth JRWP, Miller ST. Evaluation of claw toe deformity, weakness of
University (protocol # X17014). All participants signed written consent to the foot intrinsics, and posteromedial shin pain. Am J Sports Med.
participate in the study. 1989;17(6):821–7.
19. Menz HB, Zammit GV, Munteanu SE, Scott G. Plantarflexion strength of the
Consent for publication toes: age and gender differences and evaluation of a clinical screening test.
Not applicable. Foot Ankle Int. 2006;27(12):1103–8.
20. Nihal A, Goldstein J, Haas J, Hiebert R, Kummer FJ, Liederbach M,
Competing interests Trepman E. Toe flexor forces in dancers and non-dancers. Foot Ankle
The authors declare that they have no competing interests. Int. 2002;23(12):1119–23.
Bruening et al. BMC Musculoskeletal Disorders (2019) 20:608 Page 9 of 9

21. Mickle KJ, Chambers S, Steele JR, Munro BJ. A novel and reliable method to
measure toe flexor strength. Clin Biomech. 2008;23(5):683.
22. Ridge ST, Myrer JW, Olsen MT, Jurgensmeier K, Johnson AW. Reliability of
doming and toe flexion testing to quantify foot muscle strength. J Foot
Ankle Res. 2017;10(1):55.
23. Jung D-Y, Kim M-H, Koh E-K, Kwon O-Y, Cynn H-S, Lee W-H. A comparison
in the muscle activity of the abductor hallucis and the medial longitudinal
arch angle during toe curl and short foot exercises. Phys Ther Sport. 2011;
12(1):30–5.
24. Kamonseki DH, Gonçalves GA, Liu CY, Júnior IL. Effect of stretching with and
without muscle strengthening exercises for the foot and hip in patients
with plantar fasciitis: a randomized controlled single-blind clinical trial. Man
Ther. 2016;23:76–82.
25. Delwaide P, Figiel C, Richelle C. Effects of postural changes of the upper
limb on reflex transmission inthe lower limb. Cervicolumbar reflex
interactions in man. J Neurol Neurosurg Psychiatry. 1977;40(6):616–21.
26. Zehr EP, Duysens J. Regulation of arm and leg movement during human
locomotion. Neuroscientist. 2004;10(4):347–61.
27. Quek J, Treleaven J, Brauer SG, O’Leary S, Clark RA. Intra-rater reliability of
hallux flexor strength measures using the Nintendo Wii Balance Board. J
Foot Ankle Res. 2015;8(1):48.
28. Okamura K, Kanai S, Oki S, Tanaka S, Hirata N, Sakamura Y, Idemoto N, Wada
H, Otsuka A. Does the weakening of intrinsic foot muscles cause the
decrease of medial longitudinal arch height? J Phys Ther Sci. 2017;29(6):
1001–5.
29. Kendall FP, McCreary EK, Provance PG, Rodgers M, Romani WA. Muscles:
testing and function, with posture and pain (Kendall, muscles). Philadelphia:
Lippincott Williams & Wilkins; 2005.
30. Enoka RM, Christou EA, Hunter SK, Kornatz KW, Semmler JG, Taylor AM,
Tracy BL. Mechanisms that contribute to differences in motor performance
between young and old adults. J Electromyogr Kinesiol. 2003;13(1):1–12.
31. Slifkin AB, Newell KM. Noise, information transmission, and force variability. J
Exp Psychol Hum Percept Perform. 1999;25(3):837.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.