International Journal of Pediatric Otorhinolaryngology 158 (2022) 111139

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

Motion sickness and postural control among Danish adolescents

Mathias Hald a, *, Louise Devantier a, Therese Ovesen a
a
Department of Clinical Medicine, Aarhus University Hospital, Aarhus and Department of Otorhinolaryngology, Region Hospital West Jutland, Holstebro, DK-7500,
Denmark

A R T I C L E I N F O A B S T R A C T

Keywords: Objective: Easily available clinical tests to evaluate postural control are needed. Furthermore, motion sickness
Motion sickness (MS) and postural control are correlated. The aims of this study were to compare the internal validity of a set of
Balance clinical tests of postural control with the internal validity of static posturography and to evaluate possible as­
Postural control
sociations between postural control and MS.
Posturography
Methods: We included healthy subjects from a primary school in Denmark who completed questionnaires about
MS and underwent two rounds of clinical tests of postural control and static posturography using a Tetrax
Interactive Balance System two weeks apart. For clinical tests of postural control, subjects were observed for up
to 30 s standing on both legs, on one leg, on a pillow both with their eyes open and again with their eyes closed.
Results: Twenty-one subjects were included: 71% males with an average age of 13.7 years. Agreement rates
ranged from 62% to 95% between test and retest in clinical tests.; lowest for subjects standing on their non
dominant leg with their eyes open and highest for subjects performing Romberg’s test with their eyes closed
along with subjects standing on a pillow with their eyes open. For several of these tests, almost all subjects were
able to hold their balance for the full 30 s. Test-retesting using static posturography by Bland Altman plot showed
datapoints scattered equally above and below the mean line indicating no systematic bias. Results of clinical tests
and static posturography were not associated. MS was reported from 43% of subjects and a trend was observed
with high sway scores from subjects suffering from MS. This was statistically insignificant.
Conclusions: Due to a ceiling effect, subjects achieved the same scores in both rounds of testing in several of the
clinical tests, reducing the clinical importance of these tests. Compared to clinical tests, static posturography
seemingly remains the superior method when it comes to evaluation of postural control, although not as easily
applicable in a daily clinical setting. When comparing MS and postural control a trend was observed, indicating
higher sway scores in subjects suffering from MS.

1. Introduction age and postural control measured by tests assessing both spatial and
temporal dynamics of movement, though stating that these tests work in
Fast and easily applicable clinical tests applicable in bedside settings fundamentally different ways [2]. Although elaborate tests of postural
are needed for evaluation of balance and postural control among chil­ control have become increasingly cheap and easily available, they are
dren and adolescents. These would be useful in primary care as well as in not necessarily ready and available for clinical testing when needed.
specialized paediatric settings. For now, no validated clinical tests, that Therefore, simple clinical tests evaluating postural control are
do not require special equipment, are used in the clinic. warranted.
Development of motor skills depends on simultaneous development Associations between postural control and motion sickness (MS)
of postural control [1], making it important to diagnose disorders that have previously been reported [3–11]. MS may occur when either the
affect postural control. For assessment of postural control, several individual itself or its surroundings are in motion. The symptoms often
different approaches are taken. Simple clinical tests, such as Romberg, include nausea, vomiting sweating, pallor, dizziness, headache,
are often used in clinical settings. More elaborate evaluation of postural increased saliva production, malaise, drowsiness and belching. The most
control tests, static or dynamic posturography can be used along with commonly theory explaining MS is the sensory conflict theory; MS oc­
others. A study by Gagnon et al. showed a positive association between curs when an individual receive multiple conflicting sensory inputs [12,

* Corresponding author. Palle Juul-Jensens Boulevard 99, Aarhus N, 8200, Denmark.

E-mail address: [email protected] (M. Hald).

https://doi.org/10.1016/j.ijporl.2022.111139
Received 6 December 2021; Received in revised form 13 March 2022; Accepted 11 April 2022
Available online 2 May 2022
0165-5876/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
M. Hald et al. International Journal of Pediatric Otorhinolaryngology 158 (2022) 111139

13]. It is not clear why sensory conflict causes MS in some people and of signals sampled at 34 Hz. The smaller the ST, the greater the stability.
not in others. Cognition is also likely to play a role in development of Positions from the static posturography corresponded to positions in the
MS, as incoherence in motion and perceived motion has been shown to clinical tests (see above). A test-retest two weeks apart was conducted as
provoke MS [14]. well.
The highest incidence of MS has been reported for 2–12 year old Static posturography, a test in which the subject’s postural control is
children [15]. Children below the age of two years usually do not quantified, relies on measuring sway in different positions. For the
develop MS [16]. For school children aged 7–12 years the average different exercises in the test, the elements of balance such as vision and
prevalence of MS is approximately 40% with a female over­ proprioception are altered. Subjects stand on their bare feet aligned with
representation [5,7]. During adolescence the prevalence declines [17]. markings on a platform for 32 s. Different variations are used in the test.
There are no specific tests for MS. The diagnosis is based on a patient The four positions used for the static posturography were NO, NC, PO
history of recurrent symptoms in relation to episodes of mismatches and PC:
between visual and motor system stimuli, e.g. travelling in car/bus/
plane. In order to describe individual differences in MS, The Motion - Eyes open, no foam pads underneath (NO)
Sickness Susceptibility Questionnaire (MSSQ) has been used [16]. As MS - Eyes closed, no foam pads underneath (NC)
is rarely a sign of serious illness, patient history, neurological, and - Eyes open, foam pads underneath (PO)
oto-neurological examination is usually regarded sufficient in differen­ - Eyes closed, foam pads underneath (PC)
tiating between MS and disorders of the inner ear or central nervous
system. 2.3. Questionnaires
The aims of this study were to compare the internal validity of the
clinical tests with the internal validity of the static posturography and to All students were asked to fill out two questionnaires: one for de­
evaluate possible associations between postural control and MS. mographics and one for MS. The questionnaires were designed to mimic
The first hypothesis of the study was that outcomes from the clinical a typical anamnesis from a doctor, since no validated questionnaires
tests and static posturography were correlated. about motion sickness are being used in a Danish clinical setting.
The second hypothesis of the study was that subjects who reported The forms to obtain baseline demographics included age, sex,
MS had significantly different outcomes from the clinical tests and weight, height, information about sports activities, headache, migraine,
posturography than subjects without MS. previous head traumas, and medication.
In the MS demographic questionnaire all student rated their own
2. Material and methods balance from 1 to 5; previous and/or current MS; what kind of trans­
portation provoked MS (any vehicle, sailing and flying). They were also
Twenty one students from a municipal primary school in Aarhus, asked if trips above and below half an hour provoked MS; if the symp­
Denmark, were included. The included students participated in the same toms disappeared quickly thereafter, and if anyone else in their family
science class, chosen due to having about the same age as students from suffer from MS (see supplementary).
the Danish study validating static posturography for adolescents [18]. The students who reported MS, received an additional questionnaire
Parental consent was obtained from all participants. There were no in which they were asked about specific MS symptoms, and provoking/
exclusion criteria so none were excluded. relieving factors (see supplementary).
Not all questions from the questionnaire were used in the analysis.
2.1. Clinical testing
2.4. Ethics
The students were asked which leg they considered their dominant
leg e.g., which leg they would preferably use to kick a football. First, a Students included in the study remained anonymous and no personal
Romberg’s test with open eyes was performed (corresponding to the identification data were registered or obtained. Only non-invasive tests
position NO on the posturograph). Afterwards, this was repeated with not associated with pain or discomfort were used. Written consent was
closed eyes (corresponding to NC on the posturograph); on a relatively obtained for all students and parents or caregivers. The local ethical
soft pillow with open and then with closed eyes (corresponding to PO committee was contacted and stated that no approval was required.
and PC, respectively); standing on the dominant leg, and then on the
non-dominant leg with eyes open and closed. The tests were terminated 2.5. Statistical considerations
either when the students made a visible corrective movement to main­
tain balance, opened their eyes, or reached 30 s. The time was noted if Test-retest reliability was evaluated by observed agreement for
the test was terminated before they reached 30 s. clinical testing. For static posturography, Bland-Altman plots were
The results from the clinical tests were dichotomized for analysis of produced. For normally distributed data with equal variances, unpaired
test-retest; 30 s and <30 s. Test-retesting with two weeks apart was t-test was used. P-values <0.05 were considered significant. For statis­
conducted to assess the test-retest reliability of the clinical tests. For tical analysis, STATA/IC 14.2 for Mac (StataCorp, Texas, USA), was
retesting, the same observers were used for both tests. used.

2.2. Static posturography 3. Results

For comparability, static posturography was chosen due existing The study population comprised 21 students (71% male). Mean age
validation [18,19]. For the study, the Tetrax Interactive Balance Sys­ was 13.7 years and mean BMI was 19.21 (Table 1). A total of five stu­
tem™ (Tetrax IBS™) was used. This device utilizes four integrated plates dents reported daily usage of medicine (only paracetamol and allergy
that measure postural sway under different conditions. These four plates medication usage was reported). Two students had been diagnosed with
measure vertical force asserted from the anterior and posterior parts of migraine. A history of previous concussion was reported by eight
both feet. From these measurements we used the stability index (ST), students.
describing overall postural stability. The ST is derived from the formula
∑
ST = { n1[(an − na − 1)2 + (bn − bn − 1)2 + (cn − cn − 1)2 + 3.1. Clinical test of postural control
(dn − dn − 1)2}1/2/w, where a, b, c and d are the four plates measuring
force, W is the body weight, t the experimental time, and n is the number From the clinical postural control tests, results were registered from

2
M. Hald et al. International Journal of Pediatric Otorhinolaryngology 158 (2022) 111139

Table 1 Table 3
Demographic characteristics. Mean sway scores. Measurements from the first and second round of testing are
Study population (21)
denoted “1” and “2”, respectively.
Variable Mean Min Max 95% confidence Interval
Sex, n (%)
Male 15 (71) NO1ST 19.54 8 51.99 15.44 23.65
Female 6 (29) NO2ST 16.59 6.85 30.77 14.16 19.01
Age, y, SD 13.7 (0.78) NC1ST 26.29 10.19 87.08 18.62 33.97
BMI, kg/m2, SD 19.21 (3.29) NC2ST 24.75 12.64 68.4 19.36 30.15
Biking to school, yes (%) 11 (52.4) PO1ST 23.10 10.86 61.37 17.85 28.36
Physical activity, hours/week, SD 5.45 (3.90) PO2ST 22.22 9.48 50.72 17.79 26.65
PC1ST 33.75 21.42 80.35 27.20 40.31
Table 1: Data is presented as means with standard deviation (SD) or absolute
PC2ST 32.16 18.25 91.00 25.11 39.22
number with percentage.

0 to 30 s. For several of the clinical tests nearly all of the subjects were Bland Altman analysis was performed for the differences between the
able to hold their balance for the full 30 s, i.e. representing a ceiling first and second round of tests (Fig. 1). No systematic bias between first
effect (Table 2). For analysis when comparing results from the clinical and second round of testing was found.
tests to other parameters, groups were divided into two groups; scores
between 0 and 29 and 30. The results of tests on one leg with closed eyes 3.3. Questionnaire
were significantly lower than tests conducted with open eyes (t = 6,97
and t = 4.29 p < 0.0005, for tests on the dominant and non-dominant The questionnaire data revealed that 9 out of 21 (43%) reported to
leg, respectively). The two variables did not differ significantly inter­ suffer from current MS. 12.5% claimed to have suffered from MS pre­
nally. Gender, migraine status, history of previous concussions and MS viously. The student’s subjective rating of their overall balance skills
did not correlate with results of either clinical tests or posturography. No ranged from 2 to 5 on a scale from 1 to 5 with a mean of 3.63. Subjective
associations between clinical tests and weight/BMI, sports activities, balance was not associated with any measures of postural control.
subjective assessment of balance and postural control were found. In the questionnaire, the students described their current or previous
Measures from these clinical tests were not found to be associated with symptoms of MS. For each symptom, the students had to answer how
measures from static posturography. often they experienced a given symptom on a Likert scale ranging from
1: never, to 5: every time. The most commonly reported symptom of MS
3.2. Static posturography was nausea, whereas the least reported symptom was problems with
vision (Fig. 2).
Results from the static posturography are listed in Table 3. The first
and second attempts of the subjects did not differ significantly deter­ 4. Discussion
mined by paired t-test, although the mean sway values increased
insignificantly for NO, NC, PO and PC from 1st to 2nd attempt. The The primary observations of the current study were the observed
lowest mean sway score was in the NO position (19.54) and the highest agreements of the clinical tests and retests and the Bland Altman plots of
was in the PC position (33.75). The mean scores from these two were the postural control tests and retests.
significantly different (t = − 3.88, p < 0.0005) along with the difference We found high average scores in the clinical tests of postural control.
between POST and PCST (t = − 2.82, p < 0.05). When testing using The high scores in some of the clinical tests indicate the presence of a
Pearson’s r for internal correlation between the sway scores of different ceiling effect among our subjects. Due to this distribution, it was decided
positions, strong positive correlations were found for all comparisons to divide the results into two groups. For future studies, a different
(0.75 < r < 0.94, p < 0.0005). The sway scores did not differ between approach to testing should be taken, such as limiting the tests to only
genders. For these tests above, subjects were pooled for analysis dis­ standing on one leg with eyes closed and likely repeating the exercise.
regarding MS status. For NO, NC, PO and PC, the sway scores were This approach could possibly strengthen the setup and help differentiate
higher for subjects suffering from current MS compared to subjects not subjects with good postural control from subjects with inferior postural
suffering from MS, however this was not significant. To check for bias, control, although this changes the test significantly and it would likely
not be applicable in a clinical setting. Although we did not test subjects
Table 2 with verified balance disorders, this can be done in the future studies,
Clinical tests of postural control. Observed agreement is between test and retest which in turn may make it possible to identify persons with balance
listed as percentages. Corresponding positions in posturography are listed. disorders via the tests.
Exercise Mean Median Observed 95%
A strength of this study was the addition of static posturography. Due
agreement Confidence to the addition of this test, we can see that the performance of the
interval subjects in the tests of postural control remained relatively stable over
Romberg, eyes open 26.84 30 90.48 22.28 31.40 the two rounds of tests. Our results from the static posturography are in
(NO) line with the study from Kolind et al. [18], and indicate normal postural
Romberg, eyes closed 28.57 30 95.24 25.59 31.55 control among our group of subjects. Another strength of the study is the
(NC) relative comparability between the clinical tests and the static postur­
Stand on dominant leg, 26.52 30 80.95 22.83 30.21
eyes open
ography, in the sense that the positions the subjects had to take were
Stand on dominant leg, 9.75 7 90.48 6.34 13.16 similar in the clinical tests and posturography. In addition to this, the
eyes closed instruction of subjects performing the clinical tests and static postur­
Stand on non-dominant 23.38 30 61.90 18.66 28.10 ography were performed by the same group of examiners doing it
leg, eyes open
together, thus removing inter-tester variability.
Stand on non-dominant 10.76 8 80.95 6.85 14.68
leg, eyes closed The study had a few limitations, with the first one being the rela­
Stand on pillow, eyes 28.57 30 95.24 25.59 31.55 tively small number of participants. Had the number of included subjects
open (PO) been larger, it is likely that the decrease in mean sway for all or some of
Stand on pillow, eyes 28.95 30 90.48 27.44 30.47 the measurements from the static posturography from first to second
closed (PC)
round of testing would have been significant. Although insignificant, the

3
M. Hald et al. International Journal of Pediatric Otorhinolaryngology 158 (2022) 111139

Fig. 1. Bland Altman plots for NOST, NCST, POST and PCST.

approach. Instead, some studies have observed postural precursors of

MS in the temporal dynamics of movement among adults [6,9–11] and
adolescents [7,8]. When it comes to evaluating the association between
postural control and MS by qualitative tests and tests of spatial dynamics
of movement, the evidence is more conflicting [12]. This may indicate
that postural precursors of MS are best evaluated by testing of the
temporal dynamics of movement.
Being passively restrained, may in theory mitigate the relationship
between postural control and MS, however it is unclear if passive re­
straint has any positive effect on MS [12]. It should, however, be
mentioned, that passive restraint is incomplete, since passively
restrained subjects have been shown to exhibit quantifiable movements
and that they are associated to the risk of developing MS [20,21].
In conclusion, static posturography is the superior method when it
comes to evaluating postural control among healthy adolescents. More
studies on clinical testing of postural control are needed, including
testing of subjects suffering from disorders affecting postural control. We
could not show significant difference in sway or clinical tests between
subjects that did or did not suffer from MS. No associations between
Fig. 2. Frequency of the reported MS symptoms. results from clinical tests and static posturography were found.

results do suggest improvement when undergoing static posturography Financial disclosure

multiple times. This is in line with the study from Kolind et al. [19].
Another insignificant finding was the higher sway for all measurements No funding was received for this study.
from the static posturography in subjects suffering from MS. Apart from
this finding, no associations between postural control and MS were Declaration of competing interest
identified. A reason that may have contributed to our lack of significant
findings correlating MS and the clinical tests is that the subjects have The authors report no conflict of interests.
control over their own motions during the tests. Thus, it may be difficult
for any test with the subject being in control of their motions to provide
Appendix A. Supplementary data
an indication of susceptibility to MS.
Although observing a subject falling indicates insufficient postural
Supplementary data to this article can be found online at https://doi.
control, it is likely that observing the postural control is not an optimal
org/10.1016/j.ijporl.2022.111139.

4
M. Hald et al. International Journal of Pediatric Otorhinolaryngology 158 (2022) 111139

References [11] H.J. Walter, et al., Unstable coupling of body sway with imposed motion precedes
visually induced motion sickness, Hum. Mov. Sci. 64 (2019) 389–397.
[12] F.H. Previc, Intravestibular balance and motion sickness, Aerosp. Med. Hum.
[1] E. Verbecque, L. Vereeck, A. Hallemans, Postural sway in children: a literature
Perform. 89 (2) (2018) 130–140.
review, Gait Posture 49 (2016) 402–410.
[13] A. Koch, et al., The neurophysiology and treatment of motion sickness, Dtsch.
[2] I. Gagnon, B. Swaine, R. Forget, Exploring the comparability of the sensory
Ärztebl. Int. 115 (41) (2018) 687–696.
organization test and the pediatric clinical test of sensory interaction for balance in
[14] S.A.E. Nooij, et al., Beyond sensory conflict: the role of beliefs and perception in
children, Phys. Occup. Ther. Pediatr. 26 (1–2) (2006) 23–41.
motion sickness, PLoS One 16 (1) (2021), e0245295.
[3] L.J. Smart Jr., T.A. Stoffregen, B.G. Bardy, Visually induced motion sickness
[15] F. Schmal, Neuronal mechanisms and the treatment of motion sickness,
predicted by postural instability, Hum. Factors 44 (3) (2002) 451–465.
Pharmacology 91 (3–4) (2013) 229–241.
[4] B. Shahal, et al., Computerized dynamic posturography and seasickness
[16] J.F. Golding, Motion sickness susceptibility, Auton. Neurosci. 129 (1–2) (2006)
susceptibility, Laryngoscope 109 (12) (1999) 1996–2000.
67–76.
[5] I.F. Henriques, et al., Motion sickness prevalence in school children, Eur. J. Pediatr.
[17] D. Huppert, E. Grill, T. Brandt, Survey of motion sickness susceptibility in children
173 (11) (2014) 1473–1482.
and adolescents aged 3 months to 18 years, J. Neurol. 266 (Suppl 1) (2019) 65–73.
[6] T.A. Stoffregen, et al., Getting your sea legs, PLoS One 8 (6) (2013), e66949.
[18] I.K. Christensen, et al., Validation of posturographic measurements in adolescents,
[7] C.H. Chang, et al., Postural activity and motion sickness during video game play in
Otol. Neurotol. 39 (7) (2018) e568–e574.
children and adults, Exp. Brain Res. 217 (2) (2012) 299–309.
[19] I.K. Christensen, et al., Impact of a six weeks training program on postural stability
[8] C.H. Chang, et al., Control of a virtual vehicle influences postural activity and
in adolescents, Hear. Bal. Commun. 18 (3) (2020) 199–203.
motion sickness in pre-adolescent children, Hum. Mov. Sci. 78 (2021) 102832.
[20] C.H. Chang, et al., Console video games, postural activity, and motion sickness
[9] J. Munafo, M. Diedrick, T.A. Stoffregen, The virtual reality head-mounted display
during passive restraint, Exp. Brain Res. 229 (2) (2013) 235–242.
Oculus Rift induces motion sickness and is sexist in its effects, Exp. Brain Res. 235
[21] E. Faugloire, et al., Motion sickness, body movement, and claustrophobia during
(3) (2017) 889–901.
passive restraint, Exp. Brain Res. 177 (4) (2007) 520–532.
[10] F. Koslucher, J. Munafo, T.A. Stoffregen, Postural sway in men and women during
nauseogenic motion of the illuminated environment, Exp. Brain Res. 234 (9)
(2016) 2709–2720.