Journal Pre-proof

Effects of the use of anchor systems in the rehabilitation of dynamic balance and gait
in individuals with chronic dizziness of peripheral vestibular origin: A single-blinded,
randomized, controlled clinical trial

Almir Resende Coelho, Rafaela Correia Fontes, Renato Moraes, Camila de Giacomo
Carneiro Barros, Daniela Cristina Carvalho de Abreu
PII: S0003-9993(19)31003-2
DOI: https://doi.org/10.1016/j.apmr.2019.07.012
Reference: YAPMR 57647

To appear in: ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION

Received Date: 10 May 2019

Revised Date: 11 July 2019
Accepted Date: 16 July 2019

Please cite this article as: Coelho AR, Fontes RC, Moraes R, de Giacomo Carneiro Barros C, Carvalho
de Abreu DC, Effects of the use of anchor systems in the rehabilitation of dynamic balance and gait
in individuals with chronic dizziness of peripheral vestibular origin: A single-blinded, randomized,
controlled clinical trial, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2019), doi:
https://doi.org/10.1016/j.apmr.2019.07.012.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Inc. on behalf of the American Congress of Rehabilitation Medicine
Running head: Anchor systems in the rehabilitation of gait

Title: Effects of the use of anchor systems in the rehabilitation of dynamic

balance and gait in individuals with chronic dizziness of peripheral

vestibular origin: A single-blinded, randomized, controlled clinical trial

Almir Resende Coelho1,2, Rafaela Correia Fontes2, Renato Moraes3, Camila

de Giacomo Carneiro Barros1, Daniela Cristina Carvalho de Abreu1

1
Department of Health Sciences of Ribeirão Preto Medical School, University of

São Paulo, Ribeirão Preto, Brazil.

2
Department of Physiotherapy, University of Franca, Franca, Brazil.
3
School of Physical Education and Sport of Ribeirão Preto, University of São

Paulo, Ribeirão Preto, Brazil.

Acknowledgments: The authors would like to acknowledge financial support

from the Research, Teaching and Assistance Support Foundation of School of

Medicine, University of São Paulo (USP), Ribeirão Preto, SP, Brazil. They are

also grateful for the use of the Balance Master equipment of the Nucleus to

support research on chronic degenerative diseases

Disclosure statement: No potential conflict of interest was reported by the

authors.

Corresponding author: Almir Resende Coelho, Department of Health

Sciences, Ribeirão Preto School of Medicine, University of São Paulo (USP),

Ribeirão Preto, SP, Brazil; Bandeirantes Avenue, 3900; Ribeirão Preto, SP,

Brazil; CEP (Zip Code):14049-900; Telephone: +55 16 36315-4413 / 16

37118720; email: almir.resende@hotmail.com; almir.resende@unifran.edu.br

Trial Registration: Brazilian Clinical Trials Registry (REBec): RBR-2rzt5c

Research Ethics Committee: (Protocol nº 3350/2013)

Running head: Anchor systems in the rehabilitation of gait

Title: Effects of the use of anchor systems in the rehabilitation of dynamic balance

and gait in individuals with chronic dizziness of peripheral vestibular origin: A single-

blinded, randomized, controlled clinical trial

Almir Resende Coelho1,2, Rafaela Correia Fontes2, Renato Moraes3, Camila de

Giacomo Carneiro Barros1, Daniela Cristina Carvalho de Abreu1

1 Department of Health Sciences of Ribeirão Preto Medical School, University of São Paulo,

Ribeirão Preto, Brazil.

2 Department of Physiotherapy, University of Franca, Franca, Brazil.

3 School of Physical Education and Sport of Ribeirão Preto, University of São Paulo,

Ribeirão Preto, Brazil.

Acknowledgments: The authors would like to acknowledge financial support from the

Research, Teaching and Assistance Support Foundation of School of Medicine, University of

São Paulo (USP), Ribeirão Preto, SP, Brazil. They are also grateful for the use of the

Balance Master equipment of the Nucleus to support research on chronic degenerative

diseases

Disclosure statement: No potential conflict of interest was reported by the authors.

Corresponding author: Almir Resende Coelho, Department of Health Sciences, Ribeirão

Preto School of Medicine, University of São Paulo (USP), Ribeirão Preto, SP, Brazil;

Bandeirantes Avenue, 3900; Ribeirão Preto, SP, Brazil; CEP (Zip Code):14049-900;

Telephone: +55 16 36315-4413 / 16 37118720; email: almir.resende@hotmail.com;

almir.resende@unifran.edu.br

Trial Registration: Brazilian Clinical Trials Registry (REBec): RBR-2rzt5c

Research Ethics Committee: (Protocol nº 3350/2013)

1
Article Original of ARCHIVES-PMR-D-19-00556

Effects of the use of anchor systems in the rehabilitation of dynamic balance

and gait in individuals with chronic dizziness of peripheral vestibular origin: A

single-blinded, randomized, controlled clinical trial

ABSTRACT

Objective: To assess the effectiveness of the anchors in the balance rehabilitation of

participants with chronic peripheral vestibulopathy who failed to respond positively to

conventional rehabilitation for dynamic balance and gait.

Design: Assessor-blind, randomized controlled trial.

Setting: Department of Otoneurology and Laboratory of Assessment and

Rehabilitation of Equilibrium.

Participants: Women with chronic dizziness of peripheral vestibular origin (n=42),

who continued to show otoneurological symptoms for more than 6 months after

starting classic VR, with no clinical improvement observed.

Interventions: Participants were randomly assigned to receive a clinical intervention

with the anchor system, a clinical intervention without the anchor system, or no

intervention or anchor system. The intervention was based on multi-sensory

exercises for 6 weeks, twice a week, totaling 12 sessions, in groups of up to four

participants, with an average time of 40 minutes per session.

Main outcome measures and follow-up: The primary outcome was functional

balance as assessed by the short version of the Balance Evaluation Systems Test

(Mini-BESTest). The secondary outcomes were gait parameters of step width (m),

2
step length (m), and gait speed (m/sec).The measures were assessed pre- and post-

intervention, and after a 3-month follow-up period.

Results: The proposed intervention was beneficial for dizziness, balance, and gait

for both groups studied. At the 3-month follow-up, only the group that used anchors

retained the benefits related to the physical aspects of dizziness, balance, and gait.

Conclusions: The present study found that the proposed intervention protocol, with

or without the use of anchors, was beneficial for improving the dizziness, balance,

and gait. However, retention of the benefits achieved through the exercise protocol

was observed only for those using the anchor system, which promotes the use of

haptic information. The use of anchors was effective, in short protocols (12 weeks),

with maintenance of results after 3 months.

Keywords: Older adults, Dizziness, Balance, Rehabilitation, Vestibular diseases,

Anchor system

3
Abbreviations

CNS Central nervous system

DHI Dizziness handicap inventory

G1 Intervention group with anchors

G2 Intervention group without anchors

G3 Control group with vestibulopathy

Mini-BESTest Short version of the Balance Evaluation Systems Test

MoCA Montreal cognitive assessment

VR Vestibular rehabilitation

4
 Postural control enables the carrying out of functional activities and

maintenance of balance, and is based on perception-action coupling and high-level

processes of the central nervous system (CNS).1,2 The CNS also recognizes the

vestibular system as the main information source when the input provided by the

proprioceptive and visual systems is inaccurate or absent, acting as a reference

system in such situations.3

Individuals with peripheral vestibulopathies may experience chronic sensory

conflicts among the vestibular, visual, and somatosensory systems, which prevent

them from adequately modulating sensory information to maintain balance. This

conflict may cause numerous functional impairments, which reduce the ability to

carry out activities of daily living.4

Current literature indicates that classic vestibular rehabilitation (VR), based on

the protocols of Cawthorne and Cooksey, is outdated. Recent therapeutic

approaches for VR are based in four areas: gaze stability, habituation, body balance,

and gait training.5 Therefore, the search for new therapeutic approaches that

complement current VR is necessary for reducing the influence of chronic

otoneurological symptoms, which impair balance and functional performance.

A simple, inexpensive, and effective new tool to improve balance control is the

‘anchor system’.6,7 Anchors consist of a pair of cables with a small load (125 g)

attached to one end of each cable, used by a person in the same way as anchors are

used in boats. Anchors provide haptic cues about body position relative to the

5
ground, through a combination of sensory cues from skin, joint and muscle receptors,

in a single system to detect common aspects of a stimulus. Such an action-

perception system provides information about shape, texture, motion, and forces

(inertial, gravitational, and accelerative).7 Studies in different populations, including

younger and older adults with intellectual disabilities, have shown reduced postural

sway in balancing tasks 8,9,10 and a reduction in trunk velocity and acceleration during

walking, with the use of anchors.11,12,13 Of more relevance in the study of Costa et al,

2017, is the transfer of benefits with anchors, to the performance without anchors in

older adults.12 Freitas et al.13 showed that practice with anchors for a short period of

time (60 trials), performed over 2 days, was enough to observe a reduction in

postural sway, 24 h later when the balance task was performed without anchors. This

reduction in postural sway was not observed in the control group who did not use

anchors. Considering the vestibular deficit and the consequent difficulty in orienting

the head with respect to the vertical, the use of anchors can be an interesting

strategy to help improve body orientation because they provide haptic cues about the

position of the body relative to the ground.14,7 The use of anchors during VR in

individuals with chronic peripheral vestibulopathy may maximize the benefits of VR.

Thus, the purpose of this study is to assess the effectiveness of anchors in the

rehabilitation of balance, in patients diagnosed with chronic peripheral vestibulopathy

who failed to respond positively to conventional balance and gait rehabilitation.

Methods

Study design

This was a single-blinded, randomized, controlled clinical trial in which three

groups were compared, the intervention group with anchors (G1), the intervention

6
group without anchors (G2), and the control group with vestibulopathy but no

intervention or anchors (G3). The study protocol was approved by the Ethics

Committee on Human Research of the Ribeirão Preto Clinics Hospital Medicine

School, University of São Paulo, Brazil, protocol N.3350/2013, CCAE:

19716313.8.0000.5440. Written informed consent was obtained from all participants

before enrollment.

Participants

Participants were diagnosed (through patient history, caloric tests, and

computerized vector electronystagmography) by, and received care from, the

Department of Ophtalmology, Otorhinolaryngology and Head and Neck Surgery,.

School of Medicine, University of São Paulo (FMRP-USP), in the City of Ribeirão

Preto, state of São Paulo, Brazil. Only those individuals diagnosed with unilateral and

bilateral chronic peripheral vestibulopathy who did not respond positively to

conventional VR, and met the proposed eligibility criteria, were referred to the

Laboratory of Assessment and Rehabilitation of Equilibrium (LARE) to participate in

this study.

Inclusion and exclusion criteria

The inclusion criteria were (1) women; (2) those aged 60 y or older; (3) those

who previously consulted with an otorhinolaryngologist practitioner for symptoms of

dizziness over the last 6 months, and continual dizziness made worse by head

movements (indication of vestibular pathology, according to the Bárány Society),15

with a periodicity of dizziness of daily and weekly, even after having undergone VR;

and (4) those who participated in conventional VR, based on the classic protocols of

Cawthorne and Cooksey, for at least 6 months. Eligible individuals were those who

7
did not respond positively to VR, i.e., those who continued to have otoneurological

symptoms for more than 6 months after starting classical VR, with no clinical

improvement observed. The exclusion criteria were (1) medication use for dizziness

(benzodiazepines and anticonvulsants) or calcium channel blockers (cinnarizine and

flunarizine); (2) motor, visual, or cognitive limitations that would prevent performing

the assessments and proposed interventions; (3) participation in regular physical

activity two or more times per week; and (4) presence of a systemic disease not

controlled by medication. Individuals declared to be regular physical exercisers were

those who reported a 30-min session of low-level physical activity, at least twice a

week (e.g., walking, swimming, dancing, Tai Chi Chuan) or those who performed

muscle strengthening exercises, as these exercises can contribute to the

improvement of postural control and gait. Participants were asked to report the

starting of any physical exercise, or any change in their daily routine, to Collaborator

2, on a weekly basis.

The Brazilian version of the Montreal Cognitive Assessment (MoCA) was used

to evaluate cognitive aspects. The MoCA was used to exclude volunteers who

possibly had a mild cognitive disorder. The total score for this test is 30 points; a

score of 25 or more is considered normal, or not requiring other neuropsychological

testing.16

Assessment

The Dizziness Handicap Inventory (DHI) was used to assess the effects of

dizziness on activities of daily living and was self-administered in the presence of the

researcher.17 This instrument consists of 25 questions: seven items addressing

physical health, totaling 28 points; nine items addressing emotional aspects, totaling

8
36 points; and nine items considering functional capacity, totaling 36 points. The

maximum score is 100 points (the closer the score to 100, the worse the perception

of dizziness).

Primary outcome

Functional balance was adopted as the primary outcome, assessed by the

short version of the Balance Evaluation Systems Test (Mini-BESTest). This

instrument assesses the subsystems involved in balance control, and is composed of

14 items that evaluate dynamic balance,18 divided among four categories:

anticipatory postural adjustments, reactive postural control, sensory orientation, and

gait. Each item was scored from 0 to 2, with a total maximum score of 28 and a

minimum score of zero; the smaller the total score, and the scores in each domain,

the worse the functional balance. Participants performed the tasks of the Mini-

BESTest without shoes or socks.

Secondary outcome

Gait speed was adopted as the secondary outcome and was assessed by

walking on a Balance Mastera Force Plate. Participants walked on the treadmill, the

platform of which was 1.52 m long and 0.46 m wide. Three trials per participant were

performed to quantify the following variables19: step width (m) (lateral distance

between the left and right foot in successive steps, which enables calculation of the

support base during ambulation); step length (m) (longitudinal distance between the

left and right heel in successive steps); and gait speed (m/sec) (speed throughout the

activity).

9
 All tests were performed at three intervals: baseline assessment, post-

intervention assessment, and follow-up assessment 3 months after the end of the

intervention.

Sample size calculation

Sample size calculation was performed using Gpower 3.1.7 software, based

on a study by Almeida et al.20 The study surveyed normative values of clinical tests to

evaluate balance in individuals with peripheral vestibulopathy, in which the total score

of the Mini-BESTest was used as a reference for calculating sample size. We found

an effect size of 1.4019, based on a power of 0.95 and an α rate of 0.05, which

allowed for an actual power of 0.9535, with a sample of 12 participants per group, for

the use of statistical inference from a linear mixed-effects model (random and fixed

effects).

Randomization and blinding

The researcher in charge was responsible for all evaluations and was blinded

to the randomization and intervention until outcome assessments were completed.

Collaborator 1 was responsible for randomization of the samples (through random

selection in blocks of six names), and Collaborator 2 was responsible for the

therapeutic intervention process. Each participant was given a card with a numerical

identification code and was instructed to carry this identification card to every event

related to the study. The researcher in charge always identified the participant by this

card, in the baseline assessment, after 6 weeks of care (post-intervention), and after

the 3-month follow-up period. To conceal randomization, consecutively numbered,

sealed, black opaque envelopes were prepared by Collaborator 1, who had no other

involvement in the study. The envelopes were stored in a locked location and then

10
opened in sequence for each assessment to reveal the group allocation, after

finalization of data collection.

Interventions

According to the inclusion criteria, allocations were performed by an

independent collaborator who did not perform any assessments in the study.

Participants were randomized in blocks and allocated to three groups: G1 (n = 14)

intervention with anchors, G2 (n = 14) intervention without anchors, and G3 (n = 14)

control group without intervention (Figure 1).

Participants undertook two sessions per week for 6 weeks, for a total of 12

sessions, in groups of up to four participants, for 45 min per session. The therapeutic

intervention proposed in this study was based on the protocol of Coelho and Abreu21

(see supplementary material for a detailed description of the protocol), which

consisted of carrying out exercises to improve balance (i.e., single leg stance,

standing position, feet together, distribute the body weight in different directions, and

anteroposterior trunk sway), and gait training (walking on the diagonal, walking over

obstacles, walking backward, and other marching exercises). Depending on group

allocation, participants performed all exercises with anchors (G1) (holding the flexible

rod with hands in and elbows flexed between 60 and 90°, so that the rod remained

stretched without taking the weight off the floor, simulating the anchors of a ship), or

without anchors (G2). The therapeutic progression followed the specifications of the

cited protocol, which suggested one session every 2 weeks. No adverse effects were

observed during the interventions.

Statistical analyses

11
 Statistical analyses were conducted on an intention-to-treat (ITT) basis. To

ensure homogeneity among participants, profile variables were included in a simple

one-way ANOVA. In statistical analyses for comparison purposes, the linear mixed-

effects model (random and fixed effects) was used for all analyses related to

dizziness, balance, and gait. For effect size measurements of inter-group

comparisons, as well as the P value, Cohen's d was calculated. According to a study

by Hattie et al.,22 values between 0.5 and 1.0 are preferable for demonstrating the

desired effects. In the present study, moderate and large size effects were

considered as significant values for the reliability of the P value obtained from the

mixed- and random-effects model. All analyses were performed using SPSS for

Windows, version 11.0 (SPSS Inc.). Statistical significance was set at 5% (p ≤ 0.05).

Results

From a total of 75 potential participants assessed for eligibility, 25 women

were excluded by phone because they no longer felt dizzy, six did not meet the

inclusion criteria, and two declined to participate (Fig. 2). A total of 42 were

randomized to either the intervention groups (n = 28) or the control group (n = 14).

During the course of the study, four women were excluded for completing less than

20% of the interventions and two could not be evaluated in the follow-up period due

to technical problems with the Balance Master equipment.

There were no differences between groups regarding age, weight, height, BMI,

and education (Table 1), evaluated at baseline. As the population of this study was of

Brazilian nationality, the educational level of the participants was considered, to

guarantee sample homogeneity in all stages of this study. Table 2 shows the means

and standard deviations of the variables of the primary and secondary outcomes at

baseline, post-intervention, and follow-up.

12
 The DHI scores for the groups receiving interventions (G1 and G2) were

significantly lower for the functional, emotional, and overall domains (DHI total) (p <

0.01; large effect size [Cohen’s d = 1.2 ]) when compared with the control group (G3)

after the intervention (T2) and at follow-up (T3) (Table 3). G1 was the only group that

improved in the physical aspect post-intervention, and maintained this improvement

at follow-up (p < 0.001; large effect size [Cohen's d = 2.1]).

Regarding the anticipatory postural adjustment domain, only the groups that

received the interventions achieved significantly better scores (p < 0.05; large effect

size [Cohen's d = 1.1]) at post-intervention, close to the maximum score of 6. Only

the group that received the intervention with anchors (G1) maintained better

anticipatory control 3 months after the intervention when compared with the control

group (G3). There were no significant differences among groups for the reactive

postural control domain between post-intervention and follow-up.

In the sensory orientation domain, G1 and G2 improved their performances.

However, only the group that used anchors (G1) maintained a better score for the

sensory orientation domain at follow-up (p < 0.05; large effect size [Cohen's d = 1.7])

(Table 3).

For the gait domain, the intervention had a positive influence in both G1 and

G2, regardless of the use of anchors, since the scores of the two groups were

significantly higher (p < 0.05; large effect size [Cohen's d = 0.8 ]) than those of the

control group (G3).

For the total score, both intervention groups (G1 and G2) had significantly

higher scores (p < 0.05; large effect size [Cohen's d = 1.4) than the control group

13
(G3) post-intervention. However, the improvement was not retained in the follow-up

period.

As shown in Table 3, at post-intervention, the group that received intervention

with anchors (G1) showed a significantly lower step width (p < 0.001; large effect size

[Cohen's d = 1.1]) than the control group. G1 also maintained a significantly narrower

step width (p < 0.01; large effect size [Cohen's d = 0.8 ]) than the control group at

follow-up.

For step length, at post-intervention, only the group that received the

intervention with anchors (G1) had a significant increase in step length compared

with G2 and G3 (p < 0.005; moderate effect size [Cohen's d = 0.6 ]). After 3 months,

the group that used anchors (G1) maintained a longer step length than the group that

did not use anchors (G2).

At post-intervention, the group that received intervention with anchors (G1)

had a significantly faster gait speed (p < 0.05; moderate effect size [Cohen's d = 0.8

]) than G3. At follow-up, the only group that maintained a significantly faster speed (p

< 0.001; moderate effect size [Cohen's d = 0.7 ]) was the group that received

intervention with anchors (G1), compared with G2 and G3.

Discussion

The present study sought to verify the effectiveness of using a protocol of

specific exercises in combination with anchors for rehabilitating balance. The use of

anchors, as a mediator instrument of haptic cues, adds a new aspect to intervention

protocols, which explores the use of the somatosensory system. This is very

14
important because the vestibular system is deficient in individuals with peripheral

vestibulopathy and it is therefore necessary to increase the contribution of other

sensory systems to improve balance. The therapeutic method applied in this study,

using anchors as a mediator for exploring haptic cues associated with specific

physical exercises, for individuals with chronic peripheral vestibulopathy, is

unprecedented.

The results of this study showed that the proposed physical exercises

effectively improved functional and emotional aspects, as well as the total DHI scores

for the two groups that received interventions (G1 and G2). An interesting finding

related to the physical domain of the DHI was that only the group that received the

intervention with anchors improved significantly, compared with the other groups (G2

and G3) at both post-intervention and follow-up. This finding differed from those of

previous studies,23,24 in that most other researchers performed different kinds of VR

and reported reductions in scores in all aspects of the DHI. However, none of these

studies used haptic information during the VR protocol. The improvement observed

in the group that performed physical exercises with anchors was probably due to the

supplementation of haptic information generated by the anchors, which helped to

recalibrate the somatosensory system. This information improves the perception of

balance for carrying out dynamic activities, as assessed by the physical component

of the DHI.

Perceived improvement in dizziness symptoms may have been reflected in the

Mini-BESTest. Through the proposed interventions, regardless of the use of anchors,

both groups that were treated increased their scores in all aspects assessed by the

Mini-BESTest. The interventions improved anticipatory postural adjustment, gait,

15
sensory orientation, and total score, immediately after the 12 sessions. However,

only the group that used anchors was able to retain the positive results in the

domains of anticipatory postural adjustment and sensory orientation, at follow-up.

Magnani et al. (2019),25 after testing 265 seniors with the Mini-BESTest in

Brazilian communities, suggested cut-off scores for predicting falls in different age

groups of this population; a score less than or equal to 25 points could predict falls in

the 60 – 69-year-old age group. In the present study, baseline data showed that all

participants with chronic peripheral vestibular disease presented a risk of falling. After

intervention, the groups that received an intervention (G1 and G2) showed

improvements in dynamic balance and a reduced risk of falling, achieving a mean

reduction in score of 25 points when compared with G3, but did not maintain the

improvement in the follow-up period.

The improvement of dynamic balance evaluated by the Mini-BESTest was

also manifested in gait parameters observed through objective evaluation equipment.

The results showed that after intervention only G1 improved in step length, step

width, and gait speed, when compared with the other groups. Furthermore, only the

group that used anchors, concomitantly with the exercises, maintained the reduced

base of support, and increased gait speed and step length 3 months after the

intervention. These data coincide with the data from studies26,27 that assessed the

gait of older adults after VR programs and noted a reduction in step width and a

consequent reduction in the support base, which is reflected in better dynamic

balance during gait.

Kim et al.28 analyzed the gait characteristics of participants with peripheral

vestibulopathy and found that after a short walking protocol on a treadmill, there was

16
an increase in gait speed and step length. Other studies29,30 have also shown an

increase in gait speed after different VR protocols. These findings highlight the gains

obtained by individuals with chronic peripheral vestibulopathy, who were able to

improve their gait parameters, as also observed in the present study.

The present findings can be explained on the basis of studies31,7 showing that

sensory feedback is necessary for appropriate body displacement in space,

originating from the integration of visual, vestibular, and somatosensory information

that modulates the locomotor network during body displacement. This locomotor

network, as well as the CNS, is organized hierarchically so that supraspinal

locomotor centers control spinal signals. Jahn et al.31 found vestibular signs in the

locomotor areas of the somatosensory cortex, suggesting that possible alterations in

this area may cause a sensorial deficit representation of locomotor areas and

consequently promote gait disturbance.

Taking into consideration a study on the influence of vestibular information on

gait, Bent et al.32 assessed the influence of vestibular hypofunction induced by

galvanic stimulation in healthy individuals. The researchers pointed out that there

was a greater influence of vestibular information on foot placement when galvanic

stimulation was delivered during the double support phase. This finding highlights the

role of the vestibular system in planning foot placement for better gait development.

According to Mergner and Rosemeier,33 there are two modes of body segment

representation with respect to the vertical: top-down (based on vestibular input to

orient the head to the vertical) and bottom-up (based on support surface orientation

to orient the pelvis in space). Supplementation of haptic input improves body

orientation in space, using the bottom-up mode. Moreover, vestibular information

17
helps with upper segment alignment, including head alignment, during ambulation. In

the present study, only participants in the intervention group with anchors (G1) were

able to improve variables related to gait performance, and only G1 was able to

maintain a higher walking speed at follow-up, which may have occurred due to the

supplementation of haptic information. As participants with vestibulopathy have

difficulty with head orientation with respect to the vertical, anchors partially

compensate for this deficit and contribute to body orientation with respect to the

vertical, since they provide information on the body position relative to the ground.

Therefore, the improvements observed with the use of anchors is probably

due to haptic information supplementation serving as additional input to the CNS in

the process of sensory integration (mainly somatosensory information). This leads to

an improvement in the effector response pattern where vestibular sensory

information is deficient in individuals with chronic peripheral vestibulopathy.

Study limitations

The present study has some limitations. The sample did not include

participants younger than 60 years and older than 79 years, men, regular exercisers

or institutionalized older adults. Therefore, future investigations should aim to

investigate the effect of anchor systems in populations of both sexes and different

ages. In addition, we believe that the use of equipment which have platforms longer

in length than those used in the present study, could provide more detailed data on

gait.

18
Conclusions

The present study found that the proposed intervention protocol, with or

without the use of anchors, was beneficial for improving the dizziness, balance, and

gait domains evaluated in the Mini-BESTest. However, the group that underwent

intervention with anchors also showed a decrease in step width and an increase in

step length and gait speed. In a comparison of intervention groups, the use of

anchors helped to retain positive results in relation to dizziness (maintenance of the

scores related to physical domain), balance (maintenance of the scores related to

anticipatory postural adjustment and sensory orientation), and gait (maintenance of

reduced step width, increased step length, and gait speed) 3 months after the

intervention. We believe that such results were achieved due to haptic cues acting as

supplementary input to the CNS for sensory integration, to improve the effector

response pattern where there is deficient vestibular sensory information in individuals

with chronic peripheral vestibulopathy.

References

[1] M.E. Tinetti, C.S. Williams and T.M. Gill, Health, functional, and psychological

outcomes among older persons with chronic dizziness. Journal of the American

Geriatrics Society. 48 (2000), 417-421.

[2] M. A Caldas, C.F. Ganança, F.F. Ganança, M.M Ganança and H.H. Caovilla,

Vertigem posicional paroxística benigna: caracterização clínica. Braz. J

19
Otorhinolaryngol. 75 (2009), 502-6.

[3] J.M. Gazzola, F.F. Ganança, M.C. Aratani, M.R. Perracini and M.M. Ganança,

Clinical evaluation of elderly people with chronic vestibular disorder. Brazilian Journal

of Otorhinolaryngology. 72 (2006), 515-522.

[4] S.J Herdman, Vestibular Rehabilitation. 3. ed. Philadelphia:Contemporary

perspectives in rehabilitation, 2007.

[5] C.D. Hall, S.J. Herdman, S.L. Whitney, et al. Vestibular Rehabilitation for

Peripheral Vestibular Hypofunction: An Evidence-Based Clinical Practice Guideline:

FROM THE AMERICAN PHYSICAL THERAPY ASSOCIATION NEUROLOGY

SECTION. J Neurol Phys Ther. 40 (2016),124–155.

[6] E. Mauerberg-deCastro, Developing an "anchor" system to enhance postural

control. Motor Control. 8 (2004), 339-358.

[7] C. Maurer, T. Mergner, B. Bolha and F. Hlavacka, Vestibular, visual, and

somatosensory contributions to human control of upright stance. Neuroscie Lett. 281

(2000), 99-102.

[8] E. Mauerberg-deCastro, C.S. Lucena, B.W. Cuba, R.C. Boni, D.F. Campbell and

R. Moraes, Haptic stabilization of posture in adults with intellectual disabilities using a

nonrigid tool. Adapted Physical Activity Quarterly. 27 (2010),208-225.

20
[9] E. Mauerberg-deCastro, R. Moraes, D.F. Campbell. Short-term effects of the use

of non-rigid tools for postural control by adults with intellectual disabilities. Motor

Control. 16 (2012),131-143.

[10] M.B.Z. Freitas, E. Mauerberg-deCastro, R. Moraes, Intermittent use of an

"anchor system" improves postural control in healthy older adults. Gait & Posture. 38

(2013),433-437.

[11] A.A.S. Costa, P.A. Rossi, E. Mauerberg-deCastro and R. Moraes, Haptic

information provided by the "anchor system" reduces trunk sway acceleration in the

frontal plane during tandem walking in older adults. Neuroscience Letters. 609

(2015),1-6.

[12] A.A.D.S. Costa, L.O.D. Santos, E. Mauerberg-deCastro and R. Moraes, Task

difficulty has no effect on haptic anchoring during tandem walking in young and older

adults. Neuroscience Letters. 666 (2017),133-138.

[13] I. Hedayat, R. Moraes, J.L. Lanovaz and A.R. Oates, Different haptic tools

reduce trunk velocity in the frontal plane during walking, but haptic anchors have

advantages over lightly touching a railing. Experimental Brain Research. 235 (2017),

1731-1739.

[14] E. Mauerberg-deCastro, R. Moraes, C.P. Tavares, G.A. Figueiredo, S.C.M.

Pacheco and T.D.A. Costa, Haptic anchoring and human postural control.

Psychology & Neuroscience. 7 (2014), 301-318.

21
[15] Bisdorff A, Brevern MV, Lempert T, et al. Classification of vestibular symptoms:

towards an international classification of vestibular disorders. J Vestib Res. 19

(2009);1-13.

[16] Simões MR, Freitas S, Santana I, Firmino H, Martins C, Nasreddine Z, et al.

Montreal cognitive assessment (MoCA):versão final portuguesa. Coimbra: Serviço de

Avaliação Psicológica. Faculdade de Psicologia e de Ciências da Educação da

Universidade de Coimbra; 2008.

[17] Castro ASO, Gazzola JM, Natour J, Ganança FF. Versão brasileira do Dizziness

Handicap Inventory. Pró-Fono Revista de Atualização Científica. 2007;19:97-104.

[18] A.C. Maia, F. Rodrigues-de-Paula, L.C. Magalhães and R.L.L. Teixeira, Cross-

cultural adaptation and analysis of the psychometric properties of the Balance

Evaluation Systems Test and MiniBESTest in the elderly and individuals with

Parkinson's disease: application of the Rasch model. Brazilian Journal of Physical

Therapy. 17 (2013), 195-217.

[19] M. Rossi-Izquierdo, S. Santos-Perez, J.P. Rubio-Rodriguez, A. Lirola-Delgado,

A. Zubizarreta-Gutiérrez, E. San Román-Rodríguez, P. Juíz-López and A. Soto-

Varela, What is the optimal number of treatment sessions of vestibular rehabilitation?

Eur Arch Oto-Rhino-L. 271 (2014), 275–80.

[20] S.I.L Almeida, A. Marques, J. Santos. Normative values of the Balance

22
Evaluation System Test (BESTest), Mini-BESTest, Brief-BESTest, Timed Up and Go

Test and Usual Gait Speed in healthy older Portuguese people. Revista Portuguesa

de Medicina Geral e Familiar. 35 (2017), 106-116.

[21] Coelho AR, Abreu DCC. Effects of anchor system during balance rehabilitation in

subjects with chronic dizziness of peripheral vestibular origin: a controlled,

randomized, single-blind clinical trial. MOJ Gerontol Ger. 3 (2018), 414‒420. DOI:

10.15406/mojgg.2018.03.00157

[22] J. Hattie. Visible Learning. 2ª. ed. London: Routledge, 2009.

[23] G.C.S Caixeta, F. Dona, J.M. Gazzola, Cognitive processing and body balance

in elderly subjects with vestibular dysfunction. Brazilian Journal of

Otorhinolaryngology. 78 (2012), 87-95.

[24] E. Mauerberg-deCastro, R. Moraes, C.P. Tavares, G.A. Figueiredo, S.C.M.

Pacheco and T.D.A. Costa, Haptic anchoring and human postural control.

Psychology & Neuroscience. 7 (2014), 301-318.

[25] P. Magnani, M. Genovez, J. Porto, N. Zanellato, I Alvarenga, R. Freire Junior, D.

Abreu. Use of the BESTest and the Mini-BESTest for Fall Risk Prediction in

Community-Dwelling Older Adults Between 60 and 102 Years of Age. J Geriatr Phys

Ther. May (2019) Publish Ahead of Print (): DOI: 10.1519/JPT.0000000000000236.

23
[26] M. Wuehr, E. Nusser, J. Decker, S. Krafczyk, A. Straube, T. Brandt, K. Jahn and

R. Schniepp. Noisy vestibular stimulation improves dynamic walking stability in

bilateral vestibulopathy. Neurology. 86 (2016), 2196–2202.

[27] D. Meldrum, S. Herdman, R. Moloney, D. Murray, K. Malone, H. French, S.

Hone, R. Conroy and R. McConn-Walsh, Effectiveness of conventional versus virtual

reality based vestibular rehabilitation in the treatment of dizziness, gait, balance

impairment in adults with unilateral peripheral vestibular loss: a randomised

controlled trial. BMC Ear, Nose & Throat Disorders. 12 (2012), 3–8.

[28] S.C. Kim, J.Y. Kim, H.N. Lee, J.H. Kwon, N.B. Kim, M.J. Kim, J.H. Hwang and

G.C. Han, A quantitative analysis of gait patterns in vestibular neuritis patients using

gyroscope sensor and a continuous walking protocol. Journal of Neuro Engineering

and Rehabilitation. 11 (2014), 58-64.

[29] R. Schniepp, C. Schlick, F. Schenkel, C. Pradhan, K. Jahn, T. Brandt and M.

Wuehr, Clinical and neurophysiological risk factors for falls in patients with bilateral

vestibulopathy. J Neurol. 264 (2017), 277–283.

[30] M. Wuehr, R. Schniepp, C. Schlick, S. Huth, C. Pradhan, M. Dieterich, T. Brandt

and K. Jahn. Sensory loss and walking speed related factors for gait alterations in

patients with peripheral neuropathy. Gait Posture. 39 (2014), 852–858.

[31] K. Jahn, A. Deutschlander, T. Stephan, R. Kalla, K. Hüfner, J. Wagner, M.

Strupp and T. Brandt, Supraspinal locomotor control in quadrupeds and humans.

24
Prog Brain Res. 171 (2008), 353–362.

[32] L.R. Bent, J.T. Inglis, B.J McFadyen, When is vestibular information important

during walking? J Neurophysiol. 92 (2004),1269–1275.

[33] T. Mergner and T. Rosemeier, Interaction of vestibular, somatosensory and

visual signals for postural control and motion perception under terrestrial and
microgravity conditions—a conceptual model. Brain Research Reviews. 28 (1998),
118-135.

25
Figures legends

Figure 1 – Illustration of anchor system.

Figure 2 – The CONSORT flow diagram.

Supplier

a. Balance Master® platform (Neurocom International, Inc., Clackamas, OR).

26
Table 1 – Baseline demographics of participants. The values are presented as
mean ± standard deviation (SD) or as absolute number (%).

Intention – to – treat analysis (ITT)

G1 G2 G3
Variable Intervention Intervention Control group P value
with anchors without anchors (n=14)
group (n=14) group (n=14)
Age (years) 62.9 ± 1.12 63.8 ± 2.23 61.6 ± 4.14 0.324
Weight (Kg) 75.1 ± 10.3 71.4 ± 8.12 73.7 ± 9.24 0.347
Height (m) 1.60 ± 0.08 1.60 ± 0.05 1.58 ± 0.07 0.798
2
BMI (Kg/m ) 29.1 ± 2.90 28.1 ± 2.14 28.1 ± 3.87 0.617
Education (years) 8.00 ± 2.65 7.87 ± 3.47 8.16 ± 2.98 1.07
MoCA (Score) 26.8 ± 0.31 26.4± 0.98 26.8 ± 0.48 0.387

Pre-intervention participated in
conventional vestibular rehabilitation
for dizziness
Yes 14 (100%) 14 (100%) 14 (100%) 1.000
No 0 0 0

Duration of dizziness after

conventional vestibular rehabilitation 28.2 ± 2.12 26.3 ± 1.22 26.4 ± 2.43 0.687
(weeks)
Right unilateral involvement 4 (28.4%) 6 (42.6%) 6 (42.6%) 0.845
Left unilateral involvement 8 (56.8%) 7 (49.9%) 7 (49,9%) 0.947
Bilateral involvement 2 (14.2%) 1 (7.1%) 1 (7.1%) 0.897

MoCA - Montreal Cognitive Assessment. *Comparative analysis of demographic variables between groups, ANOVA
one-way, considerating p<0.05.
Table 2 – Means and Standard deviation of the variables related to the primary and secondary
outcomes of the study

Intention – to – treat analysis (ITT)

G1 G2 G3
Control group
Intervention with anchors
Measure Periods Intervention without anchors (n=14)
group (n=14)
group (n=14)
Primary outcomes
DHI (score) Baseline 56.98 ±21.27 51.23 ±26.57 53.41 ±19.98
Post Intervention 25.98 ±13.45 37.49 ±17.35 57.32 ±18.45
Follow-up 18.23 ±7.68 31.49 ±13.14 63.74 ±18.21

DHI – Physical (Score) Baseline 19.23 ±3.45 18.32 ±4.32 17.34 ±5.45
Post Intervention 9.13 ±4.23 14.75 ±5.87 19.13 ±5.74
Follow-up 8.21 ±2.74 19.32 ±6.74 23.41 ±7.32

DHI – Functional (Score) Baseline 21.23 ±7.89 18.23 ±3.23 21.58 ±8.23
Post Intervention 11.32 ±5.12 12.32 ±8.23 22.45 ±7.56
Follow-up 9.18 ±2.54 10.65 ±5.36 23.47 ±4.98

DHI – Emotional (Score) Baseline 20.14 ±2.12 18.45 ±2.78 15.84 ±7.96
Post Intervention 9.98 ±3.45 9.58 ±6.32 17.36 ±5.78
Follow-up 5.57 ±1.56 5.32 ±3.98 18.65 ±6.32

Mini BESTest ( Total Score) Baseline 22.12 ±3.78 23.68 ±2.68 22.35 ±4.36
Post Intervention 26.98 ±1.65 25.98 ±1.78 24.68 ±2.97
Follow-up 24.68 ±1.89 24.96 ±2.78 24.57 ±3.45

Mini BESTest – Anticipatory

Baseline 4.67 ±0.89 5.02 ±0.79 4.67 ±1.27
postural Adjustments (Score)
Post Intervention 5.75 ±0.35 5.23 ±0.43 4.78 ±0.83
Follow-up 5.25 ±0.78 5.12 ±0.51 4.78 ±0.78

Mini BESTest – Reactive

Baseline 4.75 ±0.75 5.28 ±0.49 4.98 ±0.97
Postural Control (Score)
Post Intervention 5.57 ±0.87 5.24 ±0.38 5.24 ±0.35
Follow-up 5.48 ±0.65 5.18 ±0.47 5.17 ±0.65

Mini BESTest - Sensory

Baseline 4.87 ±0.35 4.65 ±0.43 4.35 ±0.56
Orientation (Score)
Post Intervention 5.35 ±0.48 5.18 ±0.52 4.58 ±0.45
Follow-up 5.14 ±0.87 4.87 ±0.67 4.78 ±0.51

Mini BESTest – Gait (Score) Baseline 7.38 ±1.26 8.48 ±1.17 7.35 ±1.59
Post Intervention 9.45 ±0.89 9.75 ±0.74 8.39 ±1.37
Follow-up 8.78 ±0.98 9.06 ±0.97 8.49 ±1.23
Secondary outcomes
Step width (m) Baseline 0.17 ±0.03 0.15 ±0.08 0.15 ±0.01
Post Intervention 0.14 ±0.02 0.13 ±0.02 0.16 ±0.04
Follow-up 0.13 ±0.04 0.13 ±0.03 0.17 ±0.03

Step length (m) Baseline 0.48 ±0.14 0.49 ±0.06 0.47 ±0.30
Post Intervention 0.56 ±0.10 0.43 ±0.09 0.48 ±0.34
 Follow-up 0.53 ±0.14 0.41 ±0.08 0.51 ±0.54

Gait speed (m/sec.) Baseline 0.59 ±0.21 0.58 ±0.17 0.65 ±0.10
Post Intervention 0.68 ±0.18 0.61 ±0.13 0.59 ±0.13
Follow-up 0.61 ±0.12 0.57 ±0.17 0.57 ±0.12
Table 3 - Comparison of changes in outcome measures over time within each treatment
group.

Intention – to – treat analysis (ITT)

After 6 weeks 3 months treatment

treatment follow-up
Measure Group Mean P value P value
Difference[95%CI] Mean Difference[95%CI]
Primary outcomes
DHI (score) G1 vs. G2 -11.31[-21.45;8.41] 0.817 -13.23 [-28.41;4.78] 0.221
# #
G1 vs. G3 -31.34 [-42.32;-12.47] 0.0001 -45.51 [-53.35;-28.32] <.0001
#
G2 vs. G3 -19.83 [-33.41;-6.89] 0.005 -32.25 [-43-78;-18.63] <.0001

#
DHI – Physical (Score) G1 vs. G2 -5.62 [-10.23;-2.85] 0.058 -11.11 [-14.45;-5.32] 0.0002
# #
G1 vs. G3 -10.00 [-14.32;-5.98] <.0001 -15.20 [-18.34;-8.32] <.0001

G2 vs. G3 -4.38 [8.32;1.78] 0.387 -4.09 [-8.65;-0.13] 0.723

DHI – Functional (Score) G1 vs. G2 -3.00 [-7.65;5.63] 0.965 -1.47 [-7.98;5.65] 0.458
# #
G1 vs. G3 -11.13 [-19.47;-6.42] 0.003 -33.18 [-22.35;-8.93] <.0001
#
G2 vs. G3 -10.13 [-18.65;-6.54] 0.0003 -12.82[-9.47;-1.72] 0.073

DHI – Emotional (Score) G1 vs. G2 0.40 [-5.87;7.31] 0.963 0.25 [-6.32;5.37] 0.645
# #
G1 vs. G3 -7.38 [-12.48;-0.12] 0.052 -13.08[-19.47;-6.87] 0.001
#
G2 vs. G3 -7.78 [-14.32;-0.68] 0.033 -13.23 [-9.23;-1.67] 0.058

Mini BESTest (Score) G1 vs. G2 0.52 [-2.98;2.89] 0.745 0.13 [-3.47;1.85] 0.365
#
G1 vs. G3 -0.97 [-1.23;4.35] 0.034 0.47 [-2.85;2.98] 0.824
#
G2 vs. G3 0.45 [0.38;5.47] 0.023 0.34 [-1.98;3.47] 0.497

Mini BESTest –
Anticipatory Postural
adjustaments (Score) G1 vs. G2 0.33 [-0.87;0.98] 0.712 0.31 [-1.65;0.74] 0.247
# #
G1 vs. G3 0.33 [-0.04;1.63] 0.059 0.83 [-0.32;1.98] 0.010
#
G2 vs. G3 0.31 [-0.25;1.68] 0.031 0.47 [-0.65;1.48] 0.089

Mini BESTest – Reactive

Postural Control (Score) G1 vs. G2 0.33 [-0.35;0.64] 0.823 -0.31 [-0.78;0.35] 0.398
G1 vs. G3 0.33 [-0.09;0.78] 0.074 -0.30 [-0.68;0.45] 0.745

G2 vs. G3 0.01 [-0.18;0.97] 0.287 0.01 [-0.47;0.74] 0.629

Mini BESTest - Sensory

Orientation (Score) G1 vs. G2 0.17 [-0.56;0.74] 0.654 -0.19 [-0.87;0.08] 0.236
# #
G1 vs. G3 0.77 [0.15;1.65] 0.001 0.31 [-1.06;-0.08] 0.028
#
G2 vs. G3 0.60 [0.05;1.02] 0.047 0.09 [-0.74;0.45] 0.387

Mini BESTest – Gait (

G1 vs. G2 -0.30 [-1.35;0.87] -0.28 [-1.35;0.97]
Score) 0.498 0.947
#
G1 vs. G3 1.06 [0.32;2.87] 0.019 0.29 [-0.87;1.65] 0.365
#
G2 vs. G3 1.36 [-0.08;2.31] 0.028 0.57 [-0.68;1.98] 0.384
Secondary outcomes

Step width (m) G1 vs. G2 0.01 [-0.08;0.02] 0.378 0.00 [-0.03;0.04] 0.804
# #
G1 vs. G3 -0.02 [-0.08;-0.04] <.0001 -0.04 [-0.07;-0.02] 0.020
G2 vs. G3 -0.03 [-0.02;0.05] 0.654 -0.02 [-0.02;0.04] 0.478

# #
Step length (m) G1 vs. G2 0.13 [-0.20;0.01] 0.021 0.12 [0.01;0.28] 0.023
#
G1 vs. G3 0.08 [0.02;0.10] 0.023 0.02 [-0.03;0.17] 0.364

G2 vs. G3 -0.05 [-0.03;0.16] 0.356 -0.01 [-0.08;0.14] 0.746

#
Gait speed (m/sec.) G1 vs. G2 0.07 [-0.14;0.08] 0.987 0.14 [0.02;0.29] 0.013
# #
G1 vs. G3 0.09 [-0.02;0.21] 0.026 -0.18 [-0.28;0.05] 0.050
G2 vs. G3 0.02 [-0.03;0.21] 0.274 -0.03 [-0.15;0.09] 0.347
Supplementary material – Appendix S1

Intervention protocol

The exercises were applied to groups G1 and G2, but only group 1 used

the anchors. The exercises were performed in the sitting and standing positions.

Sitting position

• Medial-lateral body sway, weight transfer from right lower limb to the left,

increasing speed progressively;

• Perform cervical spine rotation, increasing speed;

• Stand-to-sit transfer, decreasing the base of support progressively.

Progression: 1st and 2nd weeks, 3x20 s; 3rd and 4th weeks, 3x40 s; 5th and 6th

weeks, 3x60 s.

Exercise Protocol

1) Single leg stance. Repeat on the other side. Progression: 1st and 2nd weeks,

3x10 s; 3rd and 4th weeks, 3x20 s; 5th and 6th weeks, 3x30 s.

2) Standing position, feet together, distribute the body weight in different

directions. Simulate the Stability Limit Test of the Balance Master.

(Displacement A-P; M-L; Diagonals). Progression: 1st and 2ndweeks,

distributing the body weight back and forth; 3rd and 4th weeks, adding motion to

distributing the body weight to the diagonal sides; 5th and 6th weeks, adding

motion to make circles with the body.

3)- Anteroposterior trunk sway: right foot in front of the left, distributing the body

weight back and forth, alternating foot position as the activity is completed.

Progression: 1st and 2nd weeks, 1x15 repetitions; 3rd and 4th weeks, 2x12

repetitions; 5th and 6th weeks, 3x10 repetitions.

Walking

The walking exercises were performed at a 6-meter distance, where the

progression for all groups was: 1st and 2nd weeks, 1x15 repetitions; 3rd and 4th

weeks, 2x12 repetitions; 5th and 6th weeks, 3x10 repetitions.

4) Walking with gaze deviation and fixed head: 1st and 2nd weeks, vertical

deviation; 3rd and 4th weeks, horizontal deviation; 5th and 6th weeks, deviations

in the diagonals.

5) Walking with support in the forefoot: walk on tiptoes as high as one can.

6) Walking with calcaneus support only: walk with distribution of body weight on

calcaneal.

7) Walking in the diagonal: walk in a zigzag manner.

8) Walking over obstacles: walk in a straight line and step over obstacles (shoe

boxes, cushions, etc.) – put three obstacles in a 6-meter path. From the 3rd

week: walk faster; from the 5th week: step over 2 shoe boxes (one box stacked

on the other).

9) Tandem walk: walk in a straight line with one foot in front of another.

10) Backward walking: walk in a straight line backwards.

11) Lateral walk: walk sideways to the right and to the left.
 The therapeutic progression of items 5, 6 and 7 were made based on the

increase in execution speed of the exercises weekly.

12) Walking while performing a cognitive activity: walk in a straight line and

perform some cognitive task: (1st and 3rd weeks) talk the name of animals

without repeating them, talk the name of fruits without repeating them, or talk

the names of cities, states or countries without repeating them; (4th and 6th

weeks) to reduce the base of support, as well as increase speed, as the

exercise program advances.

Adverse effects

No adverse effects were observed during the interventions.