University of Southern Denmark

The effect of long-term, group-based physical, cognitive and social activities on physical
performance in elderly, community-dwelling people with mild to moderate dementia

Junge, Tina; Knudsen, Hans Kromann; Kristensen, Hanne Kaae

Published in:
Dementia

DOI:
10.1177/1471301218806376

Publication date:
2018

Document version
Accepted manuscript

Citation for pulished version (APA):

Junge, T., Knudsen, H. K., & Kristensen, H. K. (2018). The effect of long-term, group-based physical, cognitive
and social activities on physical performance in elderly, community-dwelling people with mild to moderate
dementia. Dementia. https://doi.org/10.1177/1471301218806376

Terms of use
This work is brought to you by the University of Southern Denmark through the SDU Research Portal.
Unless otherwise specified it has been shared according to the terms for self-archiving.
If no other license is stated, these terms apply:

• You may download this work for personal use only.

• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying this open access version
If you believe that this document breaches copyright please contact us providing details and we will investigate your claim.
Please direct all enquiries to puresupport@bib.sdu.dk

Download date: 23. Sep. 2020

 The effect of long-term, group-based physical, cognitive and social activities on physical
performance in elderly, community-dwelling people with mild to moderate dementia

Tina Junge1,2, Hans Kromann Knudsen1, Hanne Kaae Kristensen1, 3

1
Health Sciences Research Centre, University College Lillebaelt, Odense, Denmark
2
Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark
3
Research Unit in Rehabilitation, Department of Clinical Research, University of Southern Denmark, Odense, Denmark

Abstract

Background Elderly people with dementia are known to be less physically active compared with
elderly, healthy people, emphasizing the need for interventions in order to maintain a high level of
independence in activities of daily living. The aim was to evaluate the effect of long-term, group-
based rehabilitation including physical activity on physical performance in elderly, community-
dwelling people with mild to moderate dementia.

Methods A quasi-experimental study of 18 elderly, community-dwelling people, diagnosed with mild

to moderate dementia, participated in an ongoing rehabilitation program based on integrated physical,
cognitive and social activities. The outcome measure was physical performance: the 30 sec sit-to-
stand test, Guralnik balance test, 10-metre walking speed test, timed 6-metre walk test, and a timed
dual task walk test. The repeated measure ANOVA was used to analyze any overall differences
between related means.

Results No significant effect of time was found for the five outcome meaures during the entire period.
The variation in the estimate of most outcome scores was higher within subjects than between subjects
during the period. Profile plots illustrated that three of the participants, who experienced severe
cognitive deterioration, markedly declined in all physical performance tests.

Conclusion The expected, progressive deterioration in physical performance was delayed in a small
group of home-dwelling people with mild to moderate dementia participating in long-term, group-
based rehabilitation. Long-term, group-based rehabilitation may have the overall potential to delay
deterioration in ADL performance in home-dwelling people with mild to moderate dementia;
however, more studies with larger samples are needed to confirm the findings of this study.

1
Background
The risk of developing dementia increases with the aging of the population, and the proportion of
people with dementia are expected to increase greatly in Denmark during the next few decades (K.
Andersen et al., 1999; Danish Dementia Research Centre, 2017), as in the rest of the world (Burge,
Kuhne, Berchtold, Maupetit, & von Gunten, 2012). Currently, 6.6% (55,000) of the Danish
population above the age of 65 have been diagnosed with dementia, with an incidence of
approximately 7,700 new cases each year (K. Andersen et al., 1999; Danish Dementia Research
Centre, 2017).

In Denmark, the government has stated that the impact of dementia must be decreased by e.g.
rehabilitation programs in order to maintain a high level of independence in Activities of Daily Living
(ADL) in order to achieve highest possible functioning in every day life as well as quality of life
(Sundhedsstyrelsen, 2016). In addition to the degree of independence in ADL and the level of physical
performance (Astell, Clark, & Hartley, 2008), other factors known of importantance for people with
dementia to maintain a high level of independence in ADL are behavioural and psychological in
nature (Gruber-Baldini, Boustani, Sloane, & Zimmerman, 2004). Also, social factors are
determinants of health, such as social support networks, support and beliefs of the family and
community (WHO, 2018).

In elderly, community-dwelling people with dementia, loss of independence in ADL is strongly

associated with reduced quality of life (C. K. Andersen, Wittrup-Jensen, Lolk, Andersen, & Kragh-
Sorensen, 2004), risk of institutionalization (Knopman et al., 1999), premature death (van Dijk et al.,
2005) and a greater burden for cohabiting family members and society (Prigerson, 2003). The decline
in ADL performance in people with dementia may not only be due to disease progression, but also to
physical inactivity associated with dementia-related apathy (Littbrand, Lundin-Olsson, Gustafson, &
Rosendahl, 2009). Elderly people with dementia are known to be less physically active compared
with elderly, healthy people, emphasizing the need for physical activity interventions for the former
(Fontaine & Haaz, 2006; Hootman et al., 2002). In general, it seems that physical activity and exercise
have the potential to slow down the development of dependence in ADL, thereby enhancing the
quality of life for people with dementia and their carers (Forbes, Forbes, Blake, Thiessen, & Forbes,
2015).

For people with dementia living in residential facilities, physical activity has been shown to have a
positive effect on overall health, physical performance and ADL (Brett, Traynor, & Stapley, 2016),
which presumably would have the same effect for community-dwelling people with dementia. When
planning physical activity interventions, consideration of content, duration, frequency and intensity
must be addressed. In a meta-analysis, it was found that a high intensity physical activity program
delayed the decline in ADL in people with moderate to severe dementia living in residential facilities
(Burge et al., 2012). Similarly, a simple exercise program of moderate intensity, for one hour twice a
week, led to a significantly slower decline in ADL score in people with dementia living in residential
facilities (Rolland et al., 2007).

2
Although there seems to be consensus regarding the effect of moderate to high intensity physical
activity on different health parameters in people with dementia living in residential facilities, evidence
of content, duration and frequency is inconclusive. This issue was highlighted in a meta-analysis,
where a large effect on physical function was seen with physical activity varying from 20 to 60
minutes per session, two to six times per week, lasting from 2 to 6 months (Heyn, Abreu, &
Ottenbacher, 2004). In another meta-analysis, reviewing the effect of physical activity on ADL
performance in people with moderate to severe dementia, an overall delay in the deterioration of ADL
performance was seen in interventions lasting from 7 to 52 weeks (Burge et al., 2012). Of the five
included studies in this meta-analysis, only one long-term study were included, reporting a significant
delay in ADL deterioration only after 12 months of program duration, with results not being
significant at the 6-month follow up (Rolland et al., 2007). Further long-term-based intervention
studies are needed in order to highlight the effect of long-term physical activity for people with
dementia, including community-dwelling people with dementia, in order to maintain a high level of
independence in ADL and to sustain quality of life.

As overall health is the result of an interaction betwen physical, mental and social components (Wade
& Halligan, 2017), physical activity must be seen as a part of a multi-component program, also
including cognitive and social components. Cognitive rehabilitation programmes that include
functional tasks are found to be promising for maintaining or improving ADL performance in people
with dementia (Wade & Halligan, 2017). The physical and mental status is also related to social
participation such as roles and interaction (WHO, 2001).

In order to adress multi-component functioning, rehabilitation as a health care process is needed.

Rehabilitation is seen as ´an educational, problem-solving process that focuses on activity limitations
and aims to opimize patient social participation and well-being, and so reduce stress on carer and or
family´ (Wade, 2005). A broad spectrum of activities containing more than one of the aforementioned
three components of physical, cognitive and social activity seems to be more beneficial than to be
engaged in only one type of activity (Karp et al., 2006).

In order to meet the multi-facetted needs of people with dementia, also with respect to the dementia-
related lack of initiative and motivation, group rehabilitation may have unique advantages beyond
those achievable through individual rehabilitation. Group rehabilitation for dementia is more
effective for improving cognitive function and global severity of dementia than
personal rehabilitation (Tanaka et al., 2017). In a recent systematic mixed studies review of the effect
and importance of physical activity on behavioural and psychological symptoms in people with
dementia, the common themes across the qualitative studies, was the importance of physical activity
being ´socially rewarding´ (T. Junge, Ahler J., Knudsen H.K., Kristensen H.K., 2018). Interaction
with others in the same situation appears to be essential to people with dementia, as they experience
a feeling of belonging (T. Junge, Ahler J., Knudsen H.K., Kristensen H.K., 2018). Similarly, the
importance of participating in a group during activity such as sports have recently been emphasized
in a Danish study of healthy, elderly people, whom expressed a higher degree of enjoyment and

3
intrinsic motivation compared to controls; mainly due to social interaction during the activity, leading
to a higher feeling of social connectedness (Pedersen et al., 2017).

Therefore, the aim of this study was to assess the effect of long-term, group-based rehabilitation
including physical activity on physical performance in elderly, community-dwelling people with mild
to moderate dementia.

Methods
Study design
This study was a quasi-experimental study, evaluating the effect of a nonrandomized, pre-post
intervention in a cohort of people with mild to moderate dementia during 9 months in the Municipality
of Svendborg, an area in the southern part of Denmark.

All participants were recruited during an ongoing, group-based rehabilitation program at the local
community centre in March 2017. Before joining the program all participants were diagnosed with
dementia at the local hospital or by their GP. Participants and relatives gave informed written consent
to participate before the study started.

The study was approved by the Danish National Committee on Health Research Ethics on 2 February
2017, and by the Danish Data Protection Agency on 14 March 2017. This study conformed with the
Declaration of Helsinki (Vollmann & Winau, 1996).

Participants
The cohort consisted of 18 elderly, community-dwelling people, diagnosed with mild to moderate
dementia. The participants were aged between 65 and 88 years, consisting of 10 men and eight women
from the Municipality of Svendborg. The exclusion criteria were moderate to severe dementia or an
inability to join a group due to e.g. agitation.

The study population was allocated into two similar groups for the program, with nine participants in
each group, as this group size was considered appropriate. All participants had followed the program
for more than half a year at the commencement of this study. Only minor changes in group
composition occurred during the intervention period due to acute events, such as illness, behavioural
disturbances or significantly decreased physical performance, e.g. due to amputation.

Purpose and content of the rehabilitation program

Health and functioning, as described by the International Classification of Functioning, Disability
and Health (ICF) is a complex and dynamic relationship of biological, psychological and social
factors highly affected by contextual factors (Wade & Halligan, 2017; WHO, 2001). Therefore,
acknowledgeging the temporal contextual of the individual and a group, it is crucial to describe the
health care setting; the context of the interventions, the organization of the service internally, and the
relation to other related services such as resources besides describing the specific actions involved
(Wade, 2005).

4
The rehabilitation program is a service offered free of charge and delivered by the Municipality of
Svendborg to elderly, community-dwelling citizens who have been diagnosed with mild to moderate
dementia. Transportation to and from the program, which took place at the training centre or at an
agreed venue, was also free of charge.

All group sessions had a duration of 2 hours, twice a week, separated by at least 2 days, except during
vacation or when the therapists were absent. Between April 2017 and November 2017, a total of 44
sessions were implemented for each participant. Every session was based on an integration of
meaningful activities comprising more than one component of physical, cognitive or social character,
acknowledging the effect of multi-component and integrated interventions.

The rehabilitation program often took place outside, at the beach, in the forest, in the garden or at a
sports venue during the spring, summer and autumn seasons. During the winter season, the group
session was often conducted indoors in a training centre. Some rehabilitation activities were carried
out in the whole group, and other activities were performed individually in the group setting, e.g. with
predetermined exercises. The physical activities had a duration of at least 45 minutes at each session.
Music accompanied most of the indoor sessions.

The combined physical, cognitive and social activities always included one or more elements of
cardiovascular exercise, walking, balance, coordination, flexibility, speed, strength, endurance
training, cognitive rehabilitation activities for memory, concentration, orientation, social interaction,
planning, organizing and cooperation. Examples of the aforementioned activities could be sit-to-stand
from a chair, dancing, walking in the forest or at the beach, playing a soccer game, kitchen activities,
singing, dancing, playing board games, having quizzes, working on Ipads and computers, painting,
telling life stories and everyday stories.

A physiotherapist and an occupational therapist, with special interest and competence in dementia,
were in charge of the program. The same two therapists conducted all group sessions.

Overall, the content of the rehabilitation program was based on recommendations from the national
Danish guidelines for dementia, the Danish Health Authority and on interventions described in other
studies (Blankevoort et al., 2010; Littbrand et al., 2009; Olazaran et al., 2010). The dosage of physical
activity was aimed to be of moderate to high intensity.

Outcome measures
The physical performance data were collected at three time points: baseline in March 2017, first
follow up in August 2017, and second follow up in November 2017. A trained team of
physiotherapists and an occupational therapist conducted the physical performance testing at the local
community training centre.

5
The primary outcome measure was physical performance, evaluated by via the 30 sec sit-to-stand
test, the Guralnik balance test, the 10-metre walking speed test, the timed 6-metre walk test, and a
timed dual task walk test.

1) The 30 sec sit-to-stand (STS) test, a validated test for evaluating lower extremity physical
performance in elderly people with functional problems, risk of fall and muscle weakness in
the legs (Schoene et al., 2013). Participants performed as many STS cycles as possible, at a
self-selected speed, beginning and ending in a sitting position, and were permitted to swing
their arms. A standard chair without arm rests was used. The number of repetitions was
counted over a 30 sec test period.

2) The Guralnik balance test is a validated, static balance test designed to test elderly people
with balance problems and risk of fall (Rossiter-Fornoff, Wolf, Wolfson, & Buchner, 1995).
The participant must be able to stand unassisted without the use of a cane or walker. First, the
participant stands with feet side by side, then in semi tandem and then in tandem. All three
components are measured over 10 seconds. The number of seconds held, if less than 10
seconds, is noted and the total score is recorded in seconds.

3) The 10-metre walking speed test is a reliabel and valid tets, measuring the walking speed over
a distance of 10 meters (Peters, Middleton, Donley, Blanck, & Fritz, 2014). The participant is
instructed to walk a 10-meter distance. Time is measured while the participant walks the
distance. The start is from zero in a static position and the participant must walk as fast as
possible. The best time of three attempts is recorded and the distance covered is divided by
the time it took the participant to walk the 10 meters.

4) The 6-min walk test is measuring the walking distance over 6 minutes and is strongly
associated with functional capacity in elderly people with moderate to severe impairment. The
test is a reliable measurement tool, capable of measuring change over time (Du, Newton,
Salamonson, Carrieri-Kohlman, & Davidson, 2009). The test is conducted on a straight 25-
meter track. Prior to commencing the walk, the participant is told that the object of the test is
to walk “as far as possible” for 6 minutes. The participant is told to walk back and forth along
the demonstrated program. The participants walked in a group of nine people at a time. They
were constantly encouraged by the therapists to maintain their speed during the walk, and
were told when they were halfway done and again when they had one minute to go. At the
end, all the participants were told to stop exactly where they were. The total distance walked
was recorded.

5) A dual task walking test is a reliable and valid test to identify older individuals at high risk of
falls (Verghese et al., 2002). Dual task walk testing is recommended for detecting risk of falls
in all cognitively impaired older adults, such as people with dementia and Alzheimer’s
disease. The dual task walking test evaluates gait speed and stride length while counting
backwards from 100 by serially subtracting sevens (Muir et al., 2012). In the current study, a

6
 modified version of the dual task walking test was applied in agreement with the therapists,
as the participants were asked to count backwards from 20 while walking a distance of 10
meters. The time taken to walk that distance was recorded.

Anthropometrics and confounder variables

Weight was measured to the nearest 0.1 kg on a portable and manual set of scales with the participant
wearing light clothes. Height was measured to the nearest 0.5 cm on a portable stadiometer at
baseline. Current medication, comorbidities and self-reported physical activity during one week were
collected in co-operation with relatives of the participants. Physical activity was self-reported in mean
hours per week, excluding the training sessions. The Mini-Mental State Examination (MMSE) was
conducted at baseline and was used to quantitatively assess the severity of cognitive impairment. The
MMSE has a satisfactory reliability, although only lower levels of sensitivity for mild degrees of
idementia was found (Tombaugh & McIntyre, 1992).

Age, sex, weight and amount of physical activity during one week were registered at baseline and at
both follow ups.

The team was thoroughly instructed and trained in all standardized test procedures before the start of
the study.

Statistics
Descriptive statistics (mean and standard deviation) were calculated for all variables.

The repeated measure ANOVA, or the mixed effects linear regression, was used to detect any overall
differences between related means for the physical performance tests, as this model takes into
consideration the interdependence between the measurements. Also, the model assesses within-
subject and between-subject variation.

The between-subject effect is the five physical performance scores at three time points tested
individually, and its error term is the subject nested in the performance score. The within-subject
factor is the time points. Its error term is the residual error for the model.

The five assumptions underpinning the one-way repeated measure ANOVA were fulfilled; the
dependent variables were measured as continuous variables. The independent variable of time
consisted of three categorical and related groups with all participants present at all three time points.
Checking the last three assumptions for the mixed effects linear regression model included checking
of the residuals, outliers, normal distribution of the dependent variable and equal variances of the
differences between all combinations of related tests.

As the repeated measure ANOVA does not allow unequal observations within subjects, profile plots
were applied by ID number to visually graph the participant’s mean score for all five performance

7
tests at the three time points. Tracking of each individual’s ID number was performed in order to
follow their individual program.

A multiple linear regression model was applied to assess the overall effect from baseline to second
follow up after 9 months. Analyses were performed on each of the five single physical performance
tests separately and controlled for the following possible confounders: sex, age, body mass index
(BMI), mean daily physical activity and MMSE score.

All calculations were performed in the statistical software program STATA IC 15 (StataCorp.
2017. Stata Statistical Software: Release 15. College Station, TX: Stata Corp LLC).

Results
Totally, 15 participants completed all three time points of testing. Only one participant dropped out
after first follow up due to severe cognitive decline. Two participants entered the study at the first
follow up.

Descriptive characteristics of the participants are summarized in Table 1. Overall, there were no
significant differences between the sexes at baseline in the five outcome measures, except for the
Guralnik Balance test, where men had a significantly higher mean score than women. In the
anthropometric data, age, height and weight were significantly higher in men than women. The
MMSE score was significantly higher in women than men.

In the mixed effect linear model, there was no significant effect of time for the five outcome meaures
during the entire period. The variation in the estimation of most outcome scores was higher within
subjects than between subjects during the period, except for the STS test, where within-subject
variation was similar to the between-subject variation (Table 2).

The profile plot illustrates the physical performance level of individual participants, exemplified by
the STS test, through the study period (Figure 1). Visual inspection of all profile plots for the five
physical performance tests revealed that three of the participants declined markedly in physical
performance during the last part of the observation period from the first follow up to the second follow
up. Subjectively, a severe cognitive decline, seen by a change in behaviour, concentration and
motivation, was observed in these participants by the therapists and the relatives. These participants
were subsequently institutionalized due to their severe cognitive decline and increased dependence
on others when performing ADL activities.

With the three participants excluded in the mixed effect linear regression analysis, the within-subject
variation was lower between the three time points, except for the Tandem test with the variation being
markedly higher. The between-subject variation was lower except for the Tandem test and the 6-min
walk test. There was a slight increase in between-subject variation in the 6-min walk test (Table 3).

8
In the multiple linear regression analysis, a small, but significant increase was seen for the STS test
(Table 4). Also, there was a slight increase in the 6-min walk test, the 10-m walking speed test and in
the Dual task test. No change over time was seen for the Tandem test (Table 4). There was no
significant effect on the estimate when adjusted for the confounders of sex, age and MMSE score for
the five performance tests. The largest effect on the estimate was seen in the 10-m walking speed test
where the effect of the confounders on the estimate increased from 0.3 sec to 0.01 sec, equivalent to
a difference in mean score walking speed of 0.05 m/s (not shown in tables).

The multiple linear regression model with the above-mentioned three participants excluded, revealed
overall minimal variations. When adjusted for the confounders of sex, age and MMSE score, the
results were the same overall, except for the 6-min walk and 10-m walk test.

Discussion
In the current study, the expected progressive deterioration in physical performance was postponed
in a small group of home communtity-dwelling people with mild to moderate dementia who
participated in long-term, group-based rehabilitation including physical activity conducted twice a
week for two hours per session during a nine-month period.

At a group level, the participants of the current study sustained their physical performance level for
almost a year. The same positive effect of physical activity on walking speed and ADL was seen in
people with dementia living in nursing homes compared with controls, when following a long-term,
moderate, multi-component program including walking, strength, balance, and flexibility training
twice a week (Rolland et al., 2007). Moreover, other studies have found that exercise interventions
have led to better physical performance, but only with a short-term intervention period (Rolland et
al., 2007). As in the study by Rolland et al., the aim of the current study was to assess the effect of an
intervention in a natural setting, but with a long-term intervention period.

Comparable to the findings of the current study, combined interventions of physical activity such as
walk, strength, balance, and flexibility training positively led to a slower decline than usal medical
care in nursing home resident people with dementia when exercising one hour twice a week during a
year (Rolland et al., 2007). Similarly, more short-term multi-component intervention studies, e.g.
combining endurance, balance and strength, demonstrated larger improvement in gait speed,
functional mobility and balance than strength training alone (Blankevoort et al., 2010). The best
results were obtained with the highest training intensity, with a duration of 12 weeks or more and a
frequency of three times a week, lasting from 45 to 60 min (Blankevoort et al., 2010). The results of
the current study identically indicate that intensity, frequency and long-term duration of a
rehabilitation program is of high importance in order to maintain or improve physical
performance.evaluating, as described by this ongoing rehabilitation program carried out two times
two hours a week, with at least 45 minutes of moderate to high physical activity intensity at every
session.

9
More systematic reviews found no consensus of solely or specific physical activity interventions,
when focusing on the positive effects on behavioural and psychological symptoms in people with
dementia (Abraha I, 2017; Barreto Pde, Demougeot, Pillard, Lapeyre-Mestre, & Rolland, 2015; Brett
et al., 2016). This may emphasize that a inter-disciplinary rehabilitation program including mental
and cognitive components as well as a broad variety of physical activity in various surroundings as
in the current study, successfully can meet the multi-facetted needs of people with dementia. Similar
results was seen with a multimodal approach combining physical and cognitive rehabilitation,
improving individual goal attainment and caregiver burden in individuals and caregivers of persons
with mild dementia (Chew, Chong, Fong, & Tay, 2015)

As rehabilitation is holistic, including multi-facetted aspects of physical, cognitive, social and

contextual character, the needs of people with dementia concerning a safe and meaningful everyday
life can be met to a larger extent than when focusing solely on one aspect. Furthermore, the high
compliance of the participants in the current study indicates that long-term, group-based rehabilitation
seem meaningful to the participants. As people with dementia may lose social routines, ensuring a
stable, externally maintained routine during a longer period can be an important part of rehabilitation
treatment (Wade, 2005), as indicated in the current study.

In the mixed effects linear regression analysis, the within-subject variation was larger than the
between-subject variation in all five outcome measures. By visual inspection of a profile plot for the
entire period, three of the participants had a clear decline in their physical performance level, as well
as a subjectively observed severe cognitive decline, which led to all three being institutionalized.
Therefore, in a similar analysis as described the three participants were excluded from the statistical
analyses, although this did not change the overall result. The finding of severe cognitive decline
followed by an acute deterioration in physical performance is similar to the results of another cohort
study, concluding that cognitive decline is strongly associated with physical decline and loss of
independence in ADL (Atkinson et al., 2005).

The STS test seemed to be the most stable test over time with within-subject variation being almost
the same as between-subject variation. A significant, but small incline in the STS score was seen
during the entire period in the multiple linear regression analysis, including only baseline and end
point measurements. The mean STS score was 10.3 in the current study including people with diverse
kinds of dementia above the age of 60, comparable to presented Danish reference values of 10 to 15
for healthy, elderly women and 11 to 17 for healthy, elderly men aged 70 to 79 (Beck, Pedersen, &
Schroll, 2005). Contrary, another Danish study found a mean baseline STS score of 12 for people
aged 50 to 79 with mild Alzheimer´s Disease, participating in moderat to high intensity aerobic
training three times a week. In a systematic review it was summarized that people with mild to
moderate dementia had a mean STS score of 8.30 (Blankevoort et al., 2010) which is remarkably
lower than the mean STS score for the group of the current study during the entire period; supporting
that long-term physical activity can maintain and also improve physical performance over time, also
in elderly people with dementia.

10
Furthermore, very small but significant effects of the 10-m walking speed test, the 6-min walk test
and the dual task were observed in the multiple linear regression analysis. Excluding the three
aforementioned participants, the group performed better in the STS test, the 10-m walking speed test,
the 6-min walk test and the Dual task test, indicating that it is possible to maintain physical
performance levels over time before a severe, cognitive decline is seen.

The strengths of this study are the prospective cohort study design in a natural experimental setting,
reflecting usual practice in a Danish municipality. Also, the long-term, rehabilitation program was
developed and implemented by the municipality and experienced therapists, combining components
of physical, cognitive as well ad social character to meet the needs of people with mild to moderate
dementia.

When interpreting the results of the current study, there are some limitations to take into account. As
the cohort consisted of only 18 participants, this could have an impact on the results and effect size
of the outcomes. Nevertheless, consistent results were found independently in all five outcome
measures, underpinning the same overall message of a delay in the progressive deterioration in
physical performance over a longer period. Although all the outcome measures were developed and
validated in elderly people, to our knowledge, none of the five outcome measures has been assessed
for responsiveness, which is why it is uncertain whether these measures are capable of detecting
relevant changes at follow ups (Connelly and Vandervoort 1997). Also, outcomes measures of ADL
and quality of life could have supported the aim of the study.

In the descpription of the intervention, the complex work of establishing and leading a group is not
accounted for. The high level of participant compliance in the group activities was undoubtedly a
result of the interaction between the multi-disciplinary therapists and the participants, as well as
between the participants, as described in detail in a qualitative study of the same cohort (T. Junge,
Knudsen, H.K., Ulrich, A., Hounsgaard, L., 2018). Also, the external validity and the replicability of
the study may be affected as the intervention did not follow a standardized protocol, but instead was
adapted to the group and the individual participants with dementia; with different activities for every
session, tailored to fit the actual requirements of the group.

In summary, the findings of the current small-scaled study add support to an emerging body of
evidence that rehabilitation including physical activity is associated with a delay in the decline in
physical performance level in people with mild to moderate dementia. Long-term, group-based
rehabilitation may have the overall potential to delay deterioration in ADL performance in home-
dwelling people with mild to moderate dementia; however, more studies with larger samples are
needed to confirm the findings of this study.

Funding acknowledgements
This work was supported by the Municipality of Svendborg and the Social, Health Care College
Funen and University College Lillebaelt.

11
Declaration of conflicting interests
The authors declare that there is no conflict of interest

References

Abraha I, R. J., Trotta FM, et al. (2017). Systematic review of systematic reviews of
nonpharmacological interventions to treat behavioural
disturbances in older patients with dementia. The SENATOR-OnTop series. BMJ Open, 7:e012759.
doi:10.1136/bmjopen-2016-012759

Andersen, C. K., Wittrup-Jensen, K. U., Lolk, A., Andersen, K., & Kragh-Sorensen, P. (2004). Ability
to perform activities of daily living is the main factor affecting quality of life in patients with
dementia. Health Qual Life Outcomes, 2, 52. doi:10.1186/1477-7525-2-52

Andersen, K., Nielsen, H., Lolk, A., Andersen, J., Becker, I., & Kragh-Sorensen, P. (1999). Incidence
of very mild to severe dementia and Alzheimer's disease in Denmark: the Odense Study. Neurology,
52(1), 85-90.

Astell, A. J., Clark, S. A., & Hartley, N. T. (2008). Predictors of discharge destination for 234 patients
admitted to a combined geriatric medicine/old age psychiatry unit. Int J Geriatr Psychiatry, 23(9),
903-908. doi:10.1002/gps.2002

Atkinson, H. H., Cesari, M., Kritchevsky, S. B., Penninx, B. W., Fried, L. P., Guralnik, J. M., &
Williamson, J. D. (2005). Predictors of combined cognitive and physical decline. J Am Geriatr Soc,
53(7), 1197-1202. doi:10.1111/j.1532-5415.2005.53362.x

Barreto Pde, S., Demougeot, L., Pillard, F., Lapeyre-Mestre, M., & Rolland, Y. (2015). Exercise
training for managing behavioral and psychological symptoms in people with dementia: A systematic
review and meta-analysis. Ageing Res Rev, 24(Pt B), 274-285. doi:10.1016/j.arr.2015.09.001

Beck, A. M., Pedersen, A. N., & Schroll, M. (2005). [Underweight and unintentional weight loss
among elderly in nursing homes and in home care--problems requiring intervention]. Ugeskr Laeger,
167(3), 272-274.

Blankevoort, C. G., van Heuvelen, M. J., Boersma, F., Luning, H., de Jong, J., & Scherder, E. J.
(2010). Review of effects of physical activity on strength, balance, mobility and ADL performance
in elderly subjects with dementia. Dement Geriatr Cogn Disord, 30(5), 392-402.
doi:10.1159/000321357

Brett, L., Traynor, V., & Stapley, P. (2016). Effects of Physical Exercise on Health and Well-Being
of Individuals Living With a Dementia in Nursing Homes: A Systematic Review. J Am Med Dir
Assoc, 17(2), 104-116. doi:10.1016/j.jamda.2015.08.016

Burge, E., Kuhne, N., Berchtold, A., Maupetit, C., & von Gunten, A. (2012). Impact of physical
activity on activity of daily living in moderate to severe dementia: a critical review. Eur Rev Aging
Phys Act, 9(1), 27-39. doi:10.1007/s11556-011-0092-y

12
Chew, J., Chong, M. S., Fong, Y. L., & Tay, L. (2015). Outcomes of a multimodal cognitive and
physical rehabilitation program for persons with mild dementia and their caregivers: a goal-oriented
approach. Clin Interv Aging, 10, 1687-1694. doi:10.2147/CIA.S93914

Danish Dementia Research Centre, D. (2017). Forekomst af demens hos ældre i Danmark. Retrieved
from http://www.videnscenterfordemens.dk/media/1269231/hele-landet-og-5-regioner-2017.pdf

Du, H., Newton, P. J., Salamonson, Y., Carrieri-Kohlman, V. L., & Davidson, P. M. (2009). A review
of the six-minute walk test: its implication as a self-administered assessment tool. Eur J Cardiovasc
Nurs, 8(1), 2-8. doi:10.1016/j.ejcnurse.2008.07.001

Fontaine, K. R., & Haaz, S. (2006). Risk factors for lack of recent exercise in adults with self-reported,
professionally diagnosed arthritis. J Clin Rheumatol, 12(2), 66-69.
doi:10.1097/01.rhu.0000208611.19231.f0

Forbes, D., Forbes, S. C., Blake, C. M., Thiessen, E. J., & Forbes, S. (2015). Exercise programs for
people with dementia. Cochrane Database Syst Rev(4), CD006489.
doi:10.1002/14651858.CD006489.pub4

Gruber-Baldini, A. L., Boustani, M., Sloane, P. D., & Zimmerman, S. (2004). Behavioral symptoms
in residential care/assisted living facilities: prevalence, risk factors, and medication management. J
Am Geriatr Soc, 52(10), 1610-1617. doi:10.1111/j.1532-5415.2004.52451.x

Heyn, P., Abreu, B. C., & Ottenbacher, K. J. (2004). The effects of exercise training on elderly
persons with cognitive impairment and dementia: a meta-analysis. Arch Phys Med Rehabil, 85(10),
1694-1704.

Hootman, J. M., Macera, C. A., Ainsworth, B. E., Addy, C. L., Martin, M., & Blair, S. N. (2002).
Epidemiology of musculoskeletal injuries among sedentary and physically active adults. Med Sci
Sports Exerc, 34(5), 838-844.

Junge, T., Ahler J., Knudsen H.K., Kristensen H.K. (2018). The effect and importance of physical
activity on behavioural and psychological symptoms in people with dementia: a systematic mixed
studies review. Dementia (London).

Junge, T., Knudsen, H.K., Ulrich, A., Hounsgaard, L. (2018). The importance of groupbased
rehabilitation in elderly home-dwelling people with dementia. Submitted. Odense, Denmark.

Karp, A., Paillard-Borg, S., Wang, H. X., Silverstein, M., Winblad, B., & Fratiglioni, L. (2006).
Mental, physical and social components in leisure activities equally contribute to decrease dementia
risk. Dement Geriatr Cogn Disord, 21(2), 65-73. doi:10.1159/000089919

Knopman, D. S., Berg, J. D., Thomas, R., Grundman, M., Thal, L. J., & Sano, M. (1999). Nursing
home placement is related to dementia progression: experience from a clinical trial. Alzheimer's
Disease Cooperative Study. Neurology, 52(4), 714-718.

Littbrand, H., Lundin-Olsson, L., Gustafson, Y., & Rosendahl, E. (2009). The effect of a high-
intensity functional exercise program on activities of daily living: a randomized controlled trial in
residential care facilities. J Am Geriatr Soc, 57(10), 1741-1749. doi:10.1111/j.1532-
5415.2009.02442.x

13
Muir, S. W., Speechley, M., Wells, J., Borrie, M., Gopaul, K., & Montero-Odasso, M. (2012). Gait
assessment in mild cognitive impairment and Alzheimer's disease: the effect of dual-task challenges
across the cognitive spectrum. Gait Posture, 35(1), 96-100. doi:10.1016/j.gaitpost.2011.08.014

Olazaran, J., Reisberg, B., Clare, L., Cruz, I., Pena-Casanova, J., Del Ser, T., . . . Muniz, R. (2010).
Nonpharmacological therapies in Alzheimer's disease: a systematic review of efficacy. Dement
Geriatr Cogn Disord, 30(2), 161-178. doi:10.1159/000316119

Pedersen, M. T., Vorup, J., Nistrup, A., Wikman, J. M., Alstrom, J. M., Melcher, P. S., . . . Bangsbo,
J. (2017). Effect of team sports and resistance training on physical function, quality of life, and
motivation in older adults. Scand J Med Sci Sports, 27(8), 852-864. doi:10.1111/sms.12823

Peters, D. M., Middleton, A., Donley, J. W., Blanck, E. L., & Fritz, S. L. (2014). Concurrent validity
of walking speed values calculated via the GAITRite electronic walkway and 3 meter walk test in the
chronic stroke population. Physiother Theory Pract, 30(3), 183-188.
doi:10.3109/09593985.2013.845805

Prigerson, H. G. (2003). Costs to society of family caregiving for patients with end-stage Alzheimer's
disease. N Engl J Med, 349(20), 1891-1892. doi:10.1056/NEJMp038157

Rolland, Y., Pillard, F., Klapouszczak, A., Reynish, E., Thomas, D., Andrieu, S., . . . Vellas, B.
(2007). Exercise program for nursing home residents with Alzheimer's disease: a 1-year randomized,
controlled trial. J Am Geriatr Soc, 55(2), 158-165. doi:10.1111/j.1532-5415.2007.01035.x

Rossiter-Fornoff, J. E., Wolf, S. L., Wolfson, L. I., & Buchner, D. M. (1995). A cross-sectional
validation study of the FICSIT common data base static balance measures. Frailty and Injuries:
Cooperative Studies of Intervention Techniques. J Gerontol A Biol Sci Med Sci, 50(6), M291-297.

Schoene, D., Wu, S. M., Mikolaizak, A. S., Menant, J. C., Smith, S. T., Delbaere, K., & Lord, S. R.
(2013). Discriminative ability and predictive validity of the timed up and go test in identifying older
people who fall: systematic review and meta-analysis. J Am Geriatr Soc, 61(2), 202-208.
doi:10.1111/jgs.12106

Sundhedsstyrelsen. (2016). Livet med demens - styrket kvalitet i indsatsen. Retrieved from
https://www.sst.dk/da/nyheder/2016/~/media/90E7F515A0684681AD101C2DAB7EEE74.ashx.

Tanaka, S., Honda, S., Nakano, H., Sato, Y., Araya, K., & Yamaguchi, H. (2017). Comparison
between group and personal rehabilitation for dementia in a geriatric health service facility: single-
blinded randomized controlled study. Psychogeriatrics, 17(3), 177-185. doi:10.1111/psyg.12212

Tombaugh, T. N., & McIntyre, N. J. (1992). The mini-mental state examination: a comprehensive
review. J Am Geriatr Soc, 40(9), 922-935.

van Dijk, P. T., Mehr, D. R., Ooms, M. E., Madsen, R., Petroski, G., Frijters, D. H., . . . Ribbe, M.
W. (2005). Comorbidity and 1-year mortality risks in nursing home residents. J Am Geriatr Soc,
53(4), 660-665. doi:10.1111/j.1532-5415.2005.53216.x

Verghese, J., Buschke, H., Viola, L., Katz, M., Hall, C., Kuslansky, G., & Lipton, R. (2002). Validity
of divided attention tasks in predicting falls in older individuals: a preliminary study. J Am Geriatr
Soc, 50(9), 1572-1576.

14
Vollmann, J., & Winau, R. (1996). Informed consent in human experimentation before the
Nuremberg code. BMJ, 313(7070), 1445-1449.

Wade, D. T. (2005). Describing rehabilitation interventions. Clin Rehabil, 19(8), 811-818.

doi:10.1191/0269215505cr923ed

Wade, D. T., & Halligan, P. W. (2017). The biopsychosocial model of illness: a model whose time
has come. Clin Rehabil, 31(8), 995-1004. doi:10.1177/0269215517709890

WHO. (2001). International Classification of Functioning, Disability and Health (ICF). Retrieved
from http://www.who.int/classifications/icf/en/

WHO. (2018). The determinants of health. Retrieved from http://www.who.int/hia/evidence/doh/en/

15
Figure 1

Figure 1. Profile plot of the participants individual course during nine months for the Sit to Stand test

16
Table 1

Table 1. Characteristics of participants presented as mean and standard deviation (SD)

All participants Female Male

(n = 18) (n = 8) (n = 10)

Age 75.9 (6.5) 74.7 (3.7) 76.8 (7.2)*

Height 171.5 (8.9) 164.8 (7.8) 176.5 (7.2)*
Weight 80.18 (12.8) 70.8 (8.4) 87.3 (11.0)*
Minimal Mental State Examination 22.51(4.5) 23.6 (4.2) 21.7 (4.6)*
Daily Physical Activity (hours) 1.37 (0.9) 1.5 (0.8) 1.3 (1.0)

Sit to Stand 10.27 (2.5) 10.4 (4.0) 10.1 (3.1)

Guralnik balance 26.64 (3.7) 25.4 (3.8) 27.6 (3.4)*

10 m walking speed 7.98 (2.8) 8.0 (1.7) 8.0 (3.4)

6 min walk 359.74 (102.2) 364.0 (71.5) 356.8 (122.4)
Dual task 9.63 (3.2) 9.0 (1.7) 10.1(3.9)

* Significant difference (p<0.05) between sexes.

17
Table 2
Table 2. Within-subject and between-subject variation of the five physical performance outcomes with confidence intervals
Physical Sit to Stand Guralnik balance 10 m walking speed 6 min walk Dual task
performance test
Within-subject 5.7 (3.5-9.6) 2.0 (0.2-2.1) 0.81 (0.5-1.3) 2837 (1727-4727) 1.8 (1.1-2.9)
Between-subject 7.0 (2.8-17.6) 12.0 (7.4-19.9) 8.71 (4.3-17.6) 8501 (3365-18470) 10.4 (5.0-21.6)

18
Table 3

Table 3. Within-subject and between-subject variation of the five physical performance outcomes with confidence intervals, with three
participants excluded
Physical Sit to Stand Guralnik balance 10 m walking speed 6 min walk Dual task
performance test
Within-subject 4.4 (2.5-7.8) 10.2 (5.9-17.6) 0.4 (0.3-0.8) 2227 (1274-3894) 0.8 (0.5-1.4)
Between-subject 7.8 (3.1-19.9) 3.5 (5.9-17.7) 10.1 (5.0-22.5) 10688 (4743-24084) 11.9 (5.5-25.6)

19
Table 4

Table 4. Multiple linear regression of the five physical performance outcomes with confidence intervals.
Model 1 Model 2
Physical performance test Coefficient p-value (95% CI) Coefficient p-value (95% CI)
Sit to Stand 0.62 0.025* (0.096-1.136) 0.76 0.039* (0.055-1.467)

Guralnik balance -0.05 0.875 (-0.734-0.637) 0.13 0.340 (-0.788-1.043)

10 m walking speed 0.99 0.002* (0.475-1.507) 1.58 0.001* (1.031- 2.134)

6 min walk 0.93 0.022* (0.169-1.690) 0.79 0.059 ( -0.039-1.635)
Dual task 0.61 0.032* (0.0679-1.159) 0.75 0.498* (-1.806-3.314)

Model 1 includes all participants, Model 2 with three participants excluded

* Significant effect (p<0.05)

20