Individualized Computer-based Training for Elderly in Nursing

Homes: A Pilot Study

Katja Orlowski, Gina Maria Gräfe, Laura Tetzlaff, Thomas Schrader and Eberhard Beck

Department of Computer Science and Media, University of Applied Sciences Brandenburg,

Magdeburger Str. 50, 14770 Brandenburg, Germany

Keywords: Balance Training, Computer-based Training, Mobility.

Abstract: In older ages, the people are affected by limitations referring to physical and cognitive functions of the body.

These limitations can lead to falls, which can be prevented by different types of physical training. Some

studies showed that different kinds of physical activity have a positive effect on the equilibrium as well as on

cognitive function. During a project an individualized computer-based training was developed. The developed

application was examined during a pilot study in a local nursing home. The results indicate that the training

intervention based on the computer-based training has a positive effect on different parameters (balance, TUG

test). The limitation of the pilot study is the small sample size, which is additionally reduced due to dropouts.

In further studies the effect of a balance training with the computer-based training will be done in other nursing

homes.

1 INTRODUCTION

The ability to keep the human body balanced is a very

complex interaction of different components: the or-

gan of equilibrium of the inner ear, the visual system

as well as the proprioceptors within the joints and

muscles ensure that the body stays upright and bal-

anced. In older ages, limitations of the sensorimotor

functions can lead to dizziness and falls (Buchner et

al., van Doorn et al.) whereby persons affected are en-

dangered to loose the ability to lead a self-determined

and thus independent life. Different studies show that

physical training has a positive effect on the general

fitness, the static and dynamic equilibrium (gait pa-

rameters) (Boa Sorte Silva et al., Hortobágy et al,

Heath et al.) as well as an impact on cognitive perfor-

mance (Huxhold et al., 2008, Didczuneit-Sandhop,

2018).

Persons aged 70 years or older have an increased

risk of getting problems with dizziness. Sixty percent

of women and 50 % of men in that age group suffer

from dizziness (Schaaf et al., 2009). In that those

cases, Schaaf et al. (2009) recommend a periodically

conducted balance training, which has a positive in-

fluence on the equilibrium and leads to prevention of

50 % of the falls.

In this paper, the pilot study including an inven-

tion utilizing an individualized computer-based ba-

lance training is presented. This paper focuses on the

impact of the balance training on the ability to keep

up or even improve the individual balance, which is

considered as static and dynamic balance.

2 MATERIALS & METHODS

The pilot study was conducted in a nursing home in

Brandenburg/Havel, which provides an accommoda-

tion for 14 elderly persons. At the starting point only

12 residents could have participated in the pilot study,

however, two of whom were not able to walk or to

stand independently since they were wheelchair us-

ers. From the remaining ten subjects six (2 m, 4 f,

mean age: 85.3 (± 5.96) years, Karnovski Index 60-

70 %, level of dizziness: 5 no, 1 light to moderate)

were willing to participate in the intervention with the

individualized computer-based training. The remain-

ing four elderly persons (1 m, 3 f, mean age: 88.3

(± 8.49) years, Karnovski Index 60 %, level of dizzi-

ness: 4 no) could be motivated to serve as control

group participating in the pre- and post-intervention

test setting. All participants or their legal representa-

tives gave written consent to participate in this study

after having been informed about the procedure, its

purpose and possible risks related to the participation.

The study was approved by the local ethics committee

664

Orlowski, K., Gräfe, G., Tetzlaff, L., Schrader, T. and Beck, E.

Individualized Computer-based Training for Elderly in Nursing Homes: A Pilot Study.

DOI: 10.5220/0009162106640669

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 5: HEALTHINF, pages 664-669

ISBN: 978-989-758-398-8; ISSN: 2184-4305

of the Brandenburg Medical School Theodor Fontane

and was carried out in line with the Declaration of

Helsinki (no. E-01-20181030).

The training intervention with the computer-based

balance training was planned to last for six weeks

with three training sessions per week. Each training

session was conducted in the same environment of the

nursing home at approximately the same time of the

day. The computer-based balance training requests

the user to move his/her center of gravity (COG)

throughout the training with the result that step by

step a hidden image becomes visible. Additionally,

music plays as long as the user moves his/her center

of pressure. Both factors, utilizing visual and auditive

stimuli, were intended to increase the test persons’

motivation, since it was shown, that the application of

socioemotionally relevant aspects result in an in-

creased effect of the respective intervention (Carsten-

sen, 2003, Mather & Carstensen, 2005).

If the extend of the motion was too small, the

music softens away before it finally stops. In that

case, the user was asked to move his/her COG again.

Figure 1: Setting of the training sessions. The elderly stands

on the force plate, which is surrounded with a safety con-

struction. A coach, nurse or caregiver supervised the elderly

during the training. During playing the elderly sees the

changes made by moving the center of gravity on the screen

and step-by-step the hidden image becomes visible.

As shown in figure 1 the elderly persons were

supervised during the training sessions by a coach

(or a nurse/caregiver).

In order to assess the impact of the balance train-

ing using the individualized computer-based training,

different tests were exerted before and after the inter-

vention. In the pretest scenario all ten test persons

took part and completed the Timed-Up-and-Go (TUG

(Podsiadlo & Richardson, 1991) test and a balance

test. Additionally, different cognitive test, such as the

Mini-Mental-State-Test (MMST (Folstein et al.,

1975)) or Trail-Making-Test (TMT, (Reitan, 1992))

were performed (results not presented).

During the TUG, the time for standing up from the

chair, walking a three meters distance, turning 180°,

going back to the chair and sitting down was regis-

tered using a stop watch. In addition, the performance

of the subject was captured with a video camera in

order to be able to evaluate the gait (dynamic balance)

of the subject.

The static balance test was conducted using a

PLUX force plate (PLUX-Wireless Biosignals S.A,

Lisbon, Portugal (Plux, 2019)) and an in-house devel-

oped software called BALANCE. The balance was

examined under four different conditions:

(1) standing on both feet with a defined distance be-

tween the feet and eyes opened,

(2) standing on both feet with a defined distance be-

tween the feet and eyes closed,

(3) semi-tandem stand with right foot in front, and

(4) semi-tandem stand with left foot in front.

Each standing position was captured for 20 seconds

and the maximum variations of the center of gravity

in the left-right (medio-lateral) as well as theback-

/forwards (anterior-posterior) direction was regis-

tered with the BALANCE software. During the bal-

ance test, a safety construction and if necessary an ex-

aminer supported the test persons in order to avoid

falls (see figure 1). Complementary to the objective

values of the COP variations, the examiners docu-

mented the level of support by using a subjective

scale: without, little, medium or large (great, major)

support.

Due to the small size of the groups (experimental

and control group), the parameters of the tests were

only considered with descriptive methods of statis-

tics. The difference of the values from pre- and post-

test settings were compared.

3 RESULTS

Training Sessions

During the six-week intervention, 18 training ses-

sions were conducted. Table 1 gives an overview on

how many sessions each subject participated in and

Individualized Computer-based Training for Elderly in Nursing Homes: A Pilot Study

665

how long the mean training sessions lasted. While

two subjects took part in almost all training sessions,

one subject missed five training sessions and three

subjects were absent for seven to eight training ses-

sions. Considering the mean time per training session,

three groups were apparent: Three subjects had a

mean session time of twelve or thirteen minutes,

while one subject played the game for seven minutes

on average while the remaining two subjects with the

lowest number of training sessions were also having

the lowest mean session times of one or two minutes.

Consequently, only three subjects were to be consid-

ered as forming the experimental group during the

further analysis.

Table 1: Number of training sessions conducted from the

six subjects during the six-week intervention. Additionally,

the mean session time in minutes is given for each subject.

Subject

Number of training

sessions

Mean session time

(min)

S01 16 13

S02 10 2

S03 11 1

S04 17 12

S05 13 12

S06 10 7

User Feedback

All the participants of the EG were motivated during

the whole intervention and gave a positive oral feed-

back to the caregiver or examiner. Surprisingly the

motivation was maintained and even increased by

changing the music and/or the images weekly as well

as choosing music and images suited to seasonal or

special events during the intervention period. The in-

itial scepticism of the participants regarding the new

and unknown technology changed to a pleasant antic-

ipation of the next training.

Balance Test

For the static balance test two different aspects had to

be considered: On the one hand, the objective meas-

urement of the variations of the COG by means of the

force platform and, on the other hand, the level of in-

dividual support documented subjectively by the ex-

aminer, which provides insight into the (potential)

improvement of physical function during the inter-

vention.

Table 2 shows the level of support of the exam-

ined subjects. The examiner used a scale of four items

to document the level of support. As it is clearly visi-

ble, there is only one subject (S01), who was able to

conduct all test settings without any help and was

standing during all test positions upright and free in

the pre- and post-intervention tests. A second subject

(S03), who could not take part in the post-intervention

test due to general health reasons, also did not need

any support during the pretest. The subjects (S04,

S05) of the EG showed a reduced level of support in

the post-intervention test in one or even all settings.

In all of the other subjects the level of support was

unchanged (S06, S09) or even increased (S07, S08)

in the post-test scenario. Furthermore, it has to be

noted that the subject S08 had to be supported

throughout the three standing positions “closed eyes”,

“tandem left” and “tandem right”. We therefore de-

cided that the data of S08 could not be considered for

further analysis, especially since the TUG test could

also not be conducted.

Table 2: Level of Support (subjective assessment of the examiner) during the static balance test conducted on the force plate

for the four standing positions. The * means that the subject have to cancel the trial before the end of the measurement was

reached.

Group Subject

Level of support

Opened Eyes Closed Eyes Tandem left Tandem right

Pre Post Pre Post Pre Post Pre Post

S01

without without without without without without without without

S04 without without medium little little little little little

S05 little without medium little large medium medium little

S07

without without medium medium little little little medium

S08 without medium medium large medium large medium large*

S09 without without little little little little little little

HEALTHINF 2020 - 13th International Conference on Health Informatics

666

Table 3: Variations of COP in anterior-posterior direction and medio-lateral direction considered as area (in mm) and given

as relative difference between pre- and posttest (in %).

Group Subject

Opened Eyes

Change of the area (%)

Closed Eyes

Change of the area (%)

Tandem left

Change of the area (%)

Tandem right

Change of the area (%)

S01 -41.0 63.3 45.8 49.5

S04 -34.7 -39.9 -85.5 -74.0

S05 -29.4 9.7 4.6 22.9

S07 -11.8 32.4 65.1 51.7

S09 -43.9 12.0 17.1 26.8

Table 4: Time needed for the task of the TUG test presented in time format (mm:ss.msms) for pretest and posttest and as

relative change between pre- and posttest (in %). Additionally, the kind of support used by the subject is given.

Group Subject

TUG time pre

(mm:ss.msms)

TUG time post

(mm:ss.msms)

Relative Change

(%)

Support during

Pre-/Posttest

S01 00:16.33 00:11.27 -31.3 No / No

S04 00:25.00 00:37.53 48.0 Walker / Crutches

S05 01:00.12 00:46.31 -21.7 Walker / Walker

S07 00:35.19 00:32.02 -8.6 Walker / Walker

S09 00:12.45 00:10.15 -15.4 No / No

In table 3 the relative changes between the results

of the pre- and post-intervention tests are given for

the area calculated from the maximum variations of

the COG in anterior-posterior and medio-lateral di-

rection. All analyzed subjects (EG: 3; CG: 2) im-

proved their balance during normal standing with

opened eyes, whereby the balance test rather deterio-

rated in subject S07 (-11.8 %) (range: -43.9 to -29.4

%).

Timed-up and Go Test

For the TUG test the individual time needed is pre-

sented in the table 4. Besides the raw time each sub-

ject needed during the pre- and post-intervention test,

the relative change in time was calculated. The calcu-

lation was based on the rounded time values and is

given as percentage. Additionally, the kind of sup-

port, which was used by each subject, is listed for the

pre- and posttest scenarios. As shown in the column

“relative change”, an improvement can be seen in

four of five subjects (EG: 2 (-31.3% to -21.7 %), CG:

2 (-15.4 % to -8.6 %)). Only the subject S04 needed a

longer time (38 vs 25 sec, deterioration: 48 %) for the

TUG test post-interventional compared to the pretest.

It has to be noted, that with exception of subject S04

all the other subjects used the same means of support

in the pre- and posttest setting.

4 DISCUSSION

As our preliminary results show no generalized con-

clusion can be drawn based on the intervention utiliz-

ing the individualized computer-based training. At

the beginning of the six-week intervention, the exper-

imental group consisted of six subjects. Due to the

fact, that three of these six subjects only attended a

small number of the training session and had a very

low mean training time, they could not be considered

as part of the experimental group in the data analysis

(balance test, TUG test). Consequently, at the end of

the pilot study only the data of three subjects who had

a considerable number of training sessions and mean

duration of the training where analyzed as experi-

mental group. Initially, four subjects were considered

as control group, only two of whom could be included

in the final analysis.

One of the most striking positive effects of the in-

tervention we would attribute to the level of support

provided by the coaches while investigating the bal-

ance. While the level of support given to the subjects

of the EG during the post-intervention test is the same

or even lower compared to the pretest setting, the sub-

jects of the control group needed at least partly more

support in the post-intervention test as compared to

the pretest.

Balance Test

Concerning the results of the balance investigations

(table 3), it can be seen that all subjects show im-

provements in the setting “eyes open”. However, only

one subject of the EG showed also a reduction in the

variation area in all the other settings (“eyes closed”,

“tandem left”, “tandem right”). Small changes in the

Individualized Computer-based Training for Elderly in Nursing Homes: A Pilot Study

667

variation area in the setting “eyes closed” and “tan-

dem left” and a moderate decline (22.9 %) can be ob-

served for a second subject of the EG. Almost the

same results were found in one other subject of the

CG with a moderate decline in the three settings “eyes

closed”, “tandem left”, “tandem right”. All of the

other subjects showed greater declines. Based on the

data of the two subjects of the EG, it has to be at least

considered that the intervention may have a positive

effect on the equilibrium of the subject, which, how-

ever, has to be proven in a larger prospective random-

ized trial.

Timed-up and Go Test

Improvements in the time needed for the TUG test

can be noticed for both groups (EG, CG). For one sub-

ject of the EG a deterioration were registered, which

can be explained with the changed walking support.

While the subject used a walker during the TUG test,

he chose crutches as support during the posttest. It is

assumed that the process of standing up from a chair

is much easier using a walker than using crutches.

Methodological Critique

The main critique is the small group of subjects in-

cluded in the study. The second problem refers to the

kind of support (wheeler, crutches) used by the sub-

ject during the TUG test. In further examinations it

should be noted, that the subject uses the same kind

of support in pre- and posttest. However, in the case

a subject used no support and walks freely in the TUG

test, and needs crutches or a walker in posttest, it

should be realized due to safety aspects.

5 CONCLUSION

Based on the very preliminary results of this pilot

study, we nonetheless feel encouraged to further in-

vestigate the effects of a computer-based balance

training on the physical and cognitive function in el-

derly people with beginning dementia. Due to the low

number of subjects included in the study and the high

rate of dropouts, the informative value remains very

low. Nevertheless, we could gain a lot of experience

conducting the test scenarios with elderly and par-

tially handicapped persons. These experiences will be

integrated in the further development and improve-

ment of the computer-based training. Furthermore,

the tests conducted in the pre- and post-intervention

test settings, will be revised critically and adjusted to

the abilities of the test persons.

ACKNOWLEDGEMENT

The authors thanks the nursing home “Geschwister

Pape GmbH” in Brandenburg/Havel for the possibil-

ity to conduct the pilot study and for all the support

given before, during and after the pilot study. Special

thanks go to the participants of the pilot study taking

part in the weekly training intervention and the other

elderly persons being part of the pilot study as mem-

bers of the control group.

Furthermore, the authors want to thank the bach-

elor student Anne Biermann, who helped to conduct

the study in the nursing home and took very good care

of the residents.

REFERENCES

Boa Sorte Silva NC, Gill DP, Gregory MA, Bocti J (2018).

Multiple-modality exercise and mind-motor training to

improve mobility in older adults: A randomized con-

trolled trial. Experimental Gerontology 103: 17-26.

Buchner DM, Larson EB. (1987). Falls and fractures in pa-

tients with Alzheimer-type dementia. Jama, 257(11),

1492-1495.

Carstensen LL, Fung HH, Charles ST: Socioemotional se-

lectivity theory and the regulation of emotion in the sec-

ond half of life. Motivation an Emotion. (2003), 27 (2),

103-123.

Didczuneit-Sandhop B (2018). Ethikantrag: Entwicklung

eines statischen und dynamischen

Gleichgewichtstrainings zur Verbesserung der

kognitiven Leistungsfähigkeit von Menschen mit

leichter und mittelschwerer Demenz.

Folstein MF, Folstein SE, Mc Mugh PR. "Mini-mental

state": a practical method for grading the cognitive state

of patients for the clinician. J Psychiatr Res (1975)

12:189-198.

Heath M, Shellington E, Titheridge S, Gill DP, Petrella RJ.

(2017). A 24-week multi-modality exercise program

improves executive control in older adults with a self-

reported cognitive complaint: Evidence from the Anti-

saccade Task. J Alzheimers Dis 56: 167-183.

Hortobágyi T, Lesinski M, Gäbler M, VanSweartingen JM,

Malatesta D, Granacher U. (2015). Effects of three

types of exercise interventions on healthy old adults’

gait speed: A systematic review and Meta-Analysis.

Sports Med 45: 1627-1643.

Huxhold O, Schäfer S, Lindenberger U. (2008).

Wechselwirkungen zwischen Sensomotorik und

Kognition im Alter. Zeitschrift für Gerontologie und

Geriatrie. doi:10.1007/s00391_008_0566_3.

Mather M, Carstensen LL: Ageing and motivated cogni-

tion: The positivity effect in attention and memory.

Trends in cognitive sciences. (2005) 9(10), 496-502.

PLUX-Wireless Biosignals, S.A. (2019). Force Platform

Data Sheet. Retrieved August 23, 2019, from

HEALTHINF 2020 - 13th International Conference on Health Informatics

668

https://plux.info/sensors/288-force-platform-8202025

04.html.

Podsiadlo D, Richardson S. The Timed Up & Go": a test of

basic functional mobility for frail elderly person. J Am

Geriatr Soc (1991) 39:142-148.

Reitan, R.M. Trail Making Test. Tucson, AZ: Reitan Neu-

ropsychology Laboratory. (1992).

Schaaf H, Walther LE, Hesse, G.: Was das Gleichgewicht

im Alter verschwinden lassen kann: Betrachtungen

über den klassischen HNO Bereich hinaus. In: HNO

kompakt (2009). http://www.drhschaaf.de/Schaaf-

SchwindelimAlterHNOcompact42009.pdf

Van Doorn C, Gruber‐Baldini A L, Zimmerman S, Richard

Hebel J, Port C L, Baumgarten M, Quinn CC, Taler G,

May C, Magaziner J. and the Epidemiology of Demen-

tia in Nursing Homes Research Group. Dementia as a

risk factor for falls and fall injuries among nursing

home residents. Journal of the American geriatrics so-

ciety. (2003) 51(9), 1213-1218.

Individualized Computer-based Training for Elderly in Nursing Homes: A Pilot Study

669