Acupressure for frail older people in community dwellings

25. Demissie S, Green RC, Mucci L et al. Reliability of informa-
tion collected by proxy in family studies of Alzheimer’s dis-
ease. Neuroepidemiology 2001; 20: 106–11.
26. Harwood RH, Foss AJE, Osborn F, Gregson RM, Zaman A,
Masud T. Falls and health status in elderly women following
first eye cataract surgery: a randomised controlled trial. Br J
Ophthalmol 2016; 89: 53–9.
27. Nunes BP, de Oliveira Saes M, Siqueira FV et al. Falls and
self-assessment of eyesight among elderly people: a
population-based study in a south Brazilian municipality.
Arch Gerontol Geriatr 2014; 59: 131–5.
28. Kvelde T, McVeigh C, Toson B et al. Depressive symptom-
atology as a risk factor for falls in older people: systematic
review and meta-analysis. J Amer Geriatric Soc 2013; 61:
694–706. doi:10.1186/1471-2318-14-120.
29. Stubbs B, Stubbs J, Gnanaraj D, Soundy A. Falls in older
adults with major depressive disorder (MDD): a systematic
review and exploratory meta-analysis of prospective studies.
Intl Psychogeriatric Assoc 2016; 28: 23–9.
30. Thurman DJ, Stevens JA, Rao JK. Practice Parameter:
Assessing patients in a neurology practice for risk of falls (an
evidence-based review). Neurology 2008; 70: 473–9.
Received 11 October 2016; editorial decision 14 February
2017

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019

Age and Ageing 2017; 46: 957–964 © The Author 2017. Published by Oxford University Press on behalf of the British Geriatrics Society.
doi: 10.1093/ageing/afx050 All rights reserved. For permissions, please email: journals.permissions@oup.com
Published electronically 2 May 2017

Acupressure for frail older people in community

dwellings—a randomised controlled trial
CLARA W. C. CHAN1, PUI HING CHAU2, ANGELA Y. M. LEUNG3,4, KWAI CHING LO1, HUAFENG SHI1,
TIN PUI YUM5, YEUK YIU LEE6, LEI LI1
1
School of Chinese Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China
2
School of Nursing, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China
3
School of Nursing, The Hong Kong Polytechnic University, Hong Kong, China
4
Centre for Gerontological Nursing, The Hong Kong Polytechnic University, Hong kong, China
5
Faculty of Education, The University of Hong Kong, Hong Kong, China
6
Yan Chai Hospital Social Services Department, Yan Chai Hospital, Hong Kong, China
Address correspondence to: L. Li. Tel: (852) 3917 6424; Fax: (852) 2872 5476. Email: llie@hku.hk

Background: ‘Frailty’ is being increasing recognised as a geriatric syndrome of growing importance in the medical field.
Acupressure is a non-pharmacological, non-invasive Traditional Chinese Medicine (TCM) treatment, which may serve to
improve the quality of life (QOL) or prevent the progressive advancement of frailty in the aged population.
Objective: to investigate the effects of a 12-week, TCM-principle guided acupressure intervention on the QOL of the frail
older people living in the community.
Methods: this is a randomised controlled trial with waitlist control design. Treatment group received 15 min of acupressure
treatment, four times a week from both TCM practitioners and trained caregivers for 12 consecutive weeks. The waitlist
control group served as a care-as-usual comparison to the treatment group for the 1st 12 weeks and then received the same
treatment. Intention-to-treat principle was followed and mixed-effects models were used for data analysis.
Results: the mean age of the participants was 76.12 ± 7.08, with a mean Tilburg frailty index score of 7.13 ± 1.76.
Significant between-group differences were found in the change of physical domain score of WHOQOL-BREF (P =
0.001); change of Pittsburgh Sleep Quality Index (P < 0.001) and pain intensity (P = 0.006) with the treatment group dem-
onstrating greater improvement. Within-group effect size analysis also indicated that the acupressure protocol has significant
impact on these areas.
Conclusions: the study’s outcomes indicated that the acupressure protocol, when applied continuously for 12 weeks, 3–4 times
a week, could improve the general QOL of frail older people living in community dwellings.

Keywords: acupressure, frailty, quality of life, traditional Chinese medicine, older people.

957
C. W. C. Chan et al.
Background
‘Frailty’ is a geriatric syndrome of growing importance in
the medical field. Large cohort studies suggest that frailty is
associated with adverse outcomes like falls, disability, hospi-
talisation, long-term care admission and mortality [1]. Its
increased prevalence in the ageing society poses a great
challenge to the public health system.
Owing to the progressive nature of frailty, researchers
suggest that proper screening should be offered in the com-
munity such that the older person at the early stages of
frailty can be identified and treated [2]. From the view of
Traditional Chinese Medicine (TCM), symptoms suffered by
frail older people could be understood as a deficiency syn-
drome. Acupressure is a non-pharmacological, non-invasive
TCM treatment which may improve the physical function
and quality of life (QOL) among frail older people. Applying
acupressure on the head and shoulder areas with a specific
technique can improve blood circulation, brighten the senses
and relieve tension. Systematic reviews also indicated that acu-
pressure could improve insomnia and reduce pain [3–8].
The acupressure protocol used in this study is designed
according to TCM principles with reference to the efficacy
observed in actual clinical practices (see Supplementary
data, available at Age and Ageing online). This 12-step proto-
col is designed to be operated by trained lay-people. Any
techniques that demand professional training or involve
potential risks are excluded from the protocol. The aim is
to investigate if such protocol can improve the QOL of frail
older people or has an impact on their physical pain, sleep
quality or psychological well-being.
Methods and implementation
This multicenter, randomised controlled trial with waiting list
control group design was conducted at four Neighbourhood
Elderly Centers (NECs) in Hong Kong from December 2014
to October 2015.
Protocol approval, registration and patient consent
The study was conducted according to the principles
expressed in the Declaration of Helsinki, approved by the
Institutional Review Board of the University of Hong Kong/
Hospital Authority Hong Kong West Cluster, and registered
at the ClinicalTrials.gov Clinical Trial Registry with the identi-
fier NCT02369094. Written informed consent was obtained
from all eligible participants.
Subjects
All participants of this study were openly recruited from four
participating NECs in Hong Kong. The participants were
required to enroled in participant–caregiver dyad, with a
designated caregiver committed to attend acupressure train-
ing and perform acupressure treatment on the participants
according to the protocol. The included participants were of
age 65 or above and scored 5 or above in the Tilburg Frailty
958
Index [9] (TFI is a validated multiple-domain screening tool
to assess frailty in older people, score ranges 0–15, score
≥5 is frail). They were also physically fit to sit on a chair and
cognitively competent to understand instructions from the
practitioner and to sign the consent form (scored 6 or above
in Abbreviated Mental Test Score [10]). Older people who
were currently receiving acupressure as a regular therapy, or
had major operations in the head, neck or shoulder area, or
had any contraindication for acupressure were excluded from
the study.
Procedure and intervention
Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019
The study adopted a waiting list control design. The waiting
list control group served as a care-as-usual comparison to the
treatment group for the 1st 12 weeks. After completing the
control group activities, participants in the waiting list control
group received the same intervention as the treatment group
after the pre-treatment data were gathered. Data of all partici-
pants who completed baseline assessments (N = 101) were
analysed for the between-group differences. The pre-treatment,
post-treatment and 3 months post-treatment data of partici-
pants who completed 80% of the treatment in both groups
(N = 79) were analysed together to investigate the efficacy of
the acupressure protocol.
Consented participant–caregiver dyads were randomly
allocated to the treatment group and waiting list control
group after all intake assessment sessions were completed.
Randomisation was done by a statistician according to a com-
puter generated random number list in blocks of 4. Four
hours’ training in two sessions were provided to caregivers of
the treatment group immediately, but not to caregivers of the
waiting list control group. The latter received their training
12 weeks later. An enroled dyad would drop out from the
study if the designated caregiver could not complete the train-
ing as scheduled or was incapable of reaching the set standard
of performing the acupressure protocol.
Participants allocated to the treatment group received
15 min acupressure treatment four times a week for 12 con-
secutive weeks, twice from a Registered Chinese Medicine
Practitioner and twice from their caregivers. Participants of
the waiting list control group received the same treatment
after completing the control group activities and assessment
(see Supplementary data, Appendix 2, available at Age and
Ageing online).
Outcome measurements and analysis
Primary outcome was the QOL of the participants against
the waiting list control group after receiving 12 weeks of
acupressure treatment, and secondary outcomes were their
psycho-social well-being; sleep quality and level of pain.
The stress level of their caregivers was also observed and
measured but will not be discussed in this article.
QOL of the participants was measured by WHOQOL-
BREF, a cross cultural instrument developed by the World
Health Organization [11]. The instrument was reported to
 Acupressure for frail older people in community dwellings
be suitable for measuring the QOL of older Chinese people
living in the community dwellings [12]. The version used in
this study was the validated version for Hong Kong
Chinese [13]. The psycho-social well-being of the partici-
pants was measured by the 15-item Geriatric Depression
Scale (GDS) [14]. The validated version in Chinese had
been used to measure psycho-social well-being of the older
people in Hong Kong in prior studies [15]. A cut-off score at
8 was established to indicate depression among the older
population in Hong Kong [16]. The change of sleep quality
was assessed by the Pittsburg Sleep Quality Index (PSQI) [17].
The Chinese version of PSQI used in this study has been con-
firmed to be a sensitive, reliable and valid outcome measuring
tool in prior studies [18]. Pain occurring frequency and pain
intensity of the participants were also measured. Pain intensity
was reported in visual analogue scale of 0–10.
SPSS version 23.0 (IBM Corp.) was used for all statistical
analysis. The intention-to-treat principle was applied to the
between-group comparison. Data of participants who com-
pleted the baseline assessment (N = 101) were all included in
the analysis. Differences of the intervention effects were
assessed using the generalised mixed models, with group
(treatment and waiting list control), time (baseline and post-
treatment) and their interaction as fixed effects; and age, gen-
der, TFI score, Abbreviated Mental Test Score (AMTS) and
Chronic diseases controlled as covariates. Random effects
were used to account for repeated measures on the same sub-
ject. SPSS procedure GENLINMIXED was used with
IDENTITY link for continuous data and with LOGIT link
for binary data. Missing data were handled by the mixed
effect models. Standardised effect size was reported in
Cohen’s d which was generated by effect size calculator [19].
For continuous data, means and standard deviations con-
verted from standard error generated by the mixed-effects
model were used. For binary data, binomial effect size display
was used. The effects of the intervention based on all parti-
cipants from treatment and waiting list control groups who
completed 80% of the acupressure treatment over time
were further investigated by generalised mixed model (as a
per-protocol analysis), with time (baseline, post-treatment
and 3 months post-treatment) as fixed effect. Cohen’s d was
employed to report the standardised effect size of the treat-
ment protocol. A P-value of less than 0.05 (two-tailed) was
considered statistically significant.
Results
One hundred and eighty-two older people showed interest in
joining the research during the recruitment period. After pre-
screening and explaining the research duration and require-
ments to the participant–caregiver dyads, we received 131
applications from the four participating NECs. One hundred
and eight participants–caregiver dyads met the inclusion cri-
teria. One hundred and six dyads agreed to sign the consent
form and were randomised into treatment group and waiting
list control group, respectively. One hundred one participants
completed the baseline assessment with 79 completed 3 months
of treatments with no less than 80% attendance and post-
treatment assessment (see Supplementary data, Appendix 3,
available at Age and Ageing online). No adverse event was
reported in the period of our study.
The mean age of the participants was 76.12 ± 7.08, with
a mean Tilburg Frailty Index score of 7.13 ± 1.76. Most of
them were living with spouse or family (75%), 18% lived
alone and the remaining with maid or others. About 84%
of the participants were reported to have chronic disease,
61% with comorbidity. Affliction by cardio-vascular disease
(including high blood pressure) was the most prevailing
condition (60%), followed by diabetes (24%). About 40%
Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019
of the participants suffered from daily pain. About 89% of
the participants reported suffering from at least one symp-
tom which might compromise their psychological well-
being. The most prevailing condition was amnesia (62%),
followed by insomnia (54%) and hypersomnia (42%).
At baseline, the age, gender, TFI score and AMTS of
the treatment and waiting list control groups were similar.
The prevalence of chronic disease was higher in the treat-
ment group. The baseline scores of various domains of
QOL were similar between groups, except that the social
domain score of the WHOQOL-BREF was lower in the
treatment group (Table 1).
After 12 weeks’ acupressure intervention, the treatment
group showed improvement in all measurements in com-
paring to the control group. (Table 2)
WHOQOL-BREF: Mixed-effects models showed that
there were significant group-by-time interactions in all four
domains of WHOQOL, indicating the change of domain
scores of the treatment group was statistically significantly dif-
ferent from that of the control group. The largest effect was
detected in the physical domain. The between-group differ-
ence in change of score was 11.74 (95% CI, 5.01 to 18.47).
However, there is no significant difference between groups in
the change of self-reported QOL and General Health.
PAIN: The pain intensity of the treatment group was
significantly reduced after treatment in comparison with the
waiting list control group (P = 0.006). The between-group
difference in change of VAS score was −1.79 (95% CI,
−3.01 to −0.52).
PSQI: The intervention had a positive impact on the sleep
quality of the participants. The reduction of PSQI score of
the treatment group was significantly more than the control
group (P < 0.001). The between-group difference in change
of score was −2.70 (95% CI, −4.13 to −1.26).
The pre-treatment, post-treatment and 3 months post-
treatment data of all 79 participants who completed 80% of
treatment sessions were analysed by mixed effect models to
further explore the efficacy of the acupressure protocol
(Table 3) (Figs. 3–7, see Supplementary data, Appendix 4,
available at Age and Ageing online).
WHOQOL-BREF: Mixed-effects analysis showed that
that there were significant improvements in physical (P <
0.001) and social (P = 0.032) domains of the participants’
QOL at post-treatment time-point. The mean score of the
physical domain increased by 8.39 (95% CI, 5.41 to 11.38)
959
C. W. C. Chan et al.

Table 1. Socio-demographic, clinical and baseline characteristics of the participants and socio-demographic characteristic of
their caregivers (Figures are mean ± SD or number (%)).
Variablea All participants N = 106b Treatment group n = 54 Control group n = 52
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Characteristics of the participants
Socio-demographic
Age 76.12 ± 7.08 (min:65, max:97) 76.33 ± 6.52 (min:65, max:89) 75.90 ± 7.68 (min:65, max:97)
Sex
Male 32 (30%) 18 (33%) 14 (27%)
Female 74 (70%) 36 (67%) 38 (73%)
Residing with
None (living alone) 19 (18%) 10 (19%) 9 (17%)
Family 79 (75%) 39 (72%) 40 (77%)

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019

Others 8 (7%) 5 (9%) 3 (6%)
Clinical
Tilburg frailty indicatorc (Scores 0–15) 7.13 ± 1.76 7.22 ± 1.78 7.04 ± 1.76
AMTSd (Scores 0–10) 8.66 ± 1.32 8.67 ± 1.24 8.65 ± 1.40
Chronic disease 89 (84%) 50 (93%)e 39 (75%)e
Pain syndrome 84 (79%) 41 (76%) 43 (83%)
WHOQOL-BREF (HK)f
Overall QOL (Scores 1–5) 3.50 ± 0.81 3.46 ± 0.89 3.55 ± 0.73
General health (Scores 1–5) 3.17 ± 0.95 3.00 ± 0.97 3.33 ± 0.91
Physical domain (Scores 0–100) 57.19 ± 17.14 53.86 ± 17.20 60.45 ± 18.42
Psychological domain (Scores 0–100) 65.36 ± 12.86 63.68 ± 14.84 67.00 ± 10.45
Social domain (Scores 0–100) 62.78 ± 11.73 59.76 ± 12.94h 65.75 ± 9.64h
Environment domain (Scores 0–100) 68.97 ± 11.43 66.90 ± 11.86 71.00 ± 10.72
Other measurements
Suffered from daily pain 40 (40%) 21 (42%) 19 (37%)
Pain intensity (Scores 0–10) 4.42 ± 2.93 4.58 ± 3.18 4.25 ± 2.69
GDS scoreg (Scores 0–15) 4.31 ± 3.65 4.64 ± 3.83 3.98 ± 3.48
PSQI scoreh (Scores 0–21) 8.76 ± 2.93 9.22 ± 4.75 8.31 ± 3.99
Suffered from poor sleep quality: PSQI≥5 80 (79%) 40 (80%) 40 (78%)
Characteristics of the caregivers
Socio-demographic
Age 62.48 ± 15.09 (min:24, max:90) 61.87 ± 16.45 (min:24, max:90) 63.12 ± 13.68 (min:29, max:84)
Sex
Male 22 (21%) 10 (19%) 12 (23%)
Female 84 (79%) 44 (81%) 40 (77%)
Relationship to the participants
Spouse 49 (46%) 24 (44%) 25 (48%)
Relatives other than spouse 14 (13%) 9 (17%) 5 (10%)
Friends and volunteers 28 (26%) 14 (26%) 14 (27%)
Paid caregiver 15 (14%) 7 (13%) 8 (15%)
a
Data are presented as mean ± SD or number (%).
b
The socio-demographic data of the elderly participants/caregivers and clinical characteristic of the participants are based on the 106 participant–caregiver dyads
recruited and randomised. Other data are based on a reduced sample size of elderly participants (N = 101, treatment group n = 50, control group n = 51) who
have completed the baseline measurements.
c
TFI is a validated multiple-domain screening tool to assess frailty. Score ≥5 is frail.
d
A lower score indicates higher cognitive impairment.
e
P-value of between-group difference P < 0.05.
f
WHOQOL-BREF (HK) = The Hong Kong Chinese version of the WHOQOL-BREF. WHOQOL-BREF (HK) consists of 2 individual items and 4 domains,
higher score indicates a better QOL.
g
GDS Score 5 or above suggests depression, Score 10 or above is always indicative of depression.
h
PSQI consists of 19 self-rated questions that are grouped into 7 component scores. The component scores sum up to a PSQI total score of 0–21, with the higher
score indicating a poorer sleep quality. A PSQI score of 5 or above indicates poor sleeping quality in general population

with a medium effect size of 0.524. There was also an
increase of score in the psychological domain; however, the
change was not statistically significant.
PAIN: Among all the individual items that showed
improvement in WHOQOL-BREF, the most significant
one was the reduction of constraints caused by physical
pain (Q3 of WHOQOL-BREF, P < 0.001). This outcome
was in line with the significant reduction of pain intensity
(P < 0.001) at post-treatment time-point; the mean score
difference between post-treatment and baseline was −1.35
(95% CI, −1.95 to −0.76) in a visual analogue scale of
0–10, with an effect size of 0.461.
GDS-15: Although score improvement of the psycho-
logical domain of WHOQOL-BREF was insignificant,

960
 Acupressure for frail older people in community dwellings

Table 2. Principal outcome analysis at 12 weeks based on intention-to-treat principle (N = 101).

Variable Treatment group Within-group Control group Within-group Between-group P-valuec
(n = 50) effect size (n = 51) effect size difference in change of
Mean ± SDa (Cohen’s d)b mean ± SDa (Cohen’s d)b score or percentage
(95% CI)
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
WHOQOL-BREF (HK)
Self-reported overall QOLd (Scores 1–5)
Baseline 3.49 ± 0.78 3.54 ± 0.79
12 weeks 3.70 ± 0.64 0.294 3.45 ± 0.71 −0.120 0.30 (−0.06 to 0.67) 0.105
Self-reported General Healthe (Scores 1–5)
Baseline 3.05 ± 0.92 3.30 ± 0.93
12 weeks 3.23 ± 1.06 0.181 3.10 ± 1.14 −0.192 0.37 (−0.12 to 0.87) 0.140
Physical domain (Scores 0–100)

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019

Baseline 55.03 ± 14.42 59.79 ± 14.35
12 weeks 63.49 ± 15.69 0.561 56.52 ± 16.71 −0.210 11.74 (5.01 to 18.47) 0.001*
Psychological domain (Scores 0–100)
Baseline 63.97 ± 12.02 66.90 ± 12.00
12 weeks 67.91 ± 12.94 0.315 64.89 ± 13.78 −0.156 5.95 (0.20 to 11.70) 0.043*
Social domain (Scores 0–100)
Baseline 59.89 ± 11.31 65.69 ± 11.28
12 weeks 64.45 ± 10.47 0.418 63.70 ± 11.21 −0.177 6.54 (1.40 to 11.69) 0.013*
Environment domain (Scores 0–100)
Baseline 67.31 ± 10.89 70.78 ± 10.85
12 weeks 71.61 ± 11.81 0.379 68.13 ± 12.57 −0.226 6.95 (2.02 to 11.87) 0.006*
Suffered from daily pain
Baseline 42% 37%
12 weeks 30% −0.243 52% 0.366 −25% (−46% to −12%) 0.022*
Pain Intensity (Scores 0–10)
Baseline 4.60 ± 2.76 4.23 ± 2.79
12 weeks 3.24 ± 3.04 −0.468 4.67 ± 3.29 0.144 −1.79 (−3.01 to −0.52) 0.006*
GDS (Scores 0–15)f
Baseline 4.53 ± 3.32 4.01 ± 3.29
12 weeks 3.39 ± 3.46 −0.336 4.26 ± 3.64 0.072 −1.39 (−2.68 to −0.09) 0.036*
PSQI (Scores 0–21)g
Baseline 9.31 ± 4.31 8.21 ± 4.28
12 weeks 7.83 ± 4.38 −0.341 9.42 ± 4.57 0.273 −2.70 (−4.13 to −1.26) <0.001*

*P < 0.05.
a
The estimated mean from linear mixed-effects model with age; gender; TFI score; AMTS and Chronic diseases controlled as covariates, SD converted from stand-
ard error generated from the same model.
b
Cohen’s d of 0.2 represents a small effect size, 0.5 a medium effect size and 0.8 a large effect size. Cohen’s d is generated by effect size calculator (https://www.
psychometrica.de/effect_size.html). For continuous variable, means and standard deviations converted from standard error generated by the mixed model were
used as input. For binary data, estimated number of success and failure generated by the mixed model were used as input.
c
P-value for group-by-time interaction based on generalised mixed-effects model is generated by SPSS procedure: GENLINMIXED. For continuous variables, an
IDENTITY link is used; for binary variable, a LOGIT link is used.
d
Score of WHOQOL-BREF Question 1.
e
Score of WHOQOL-BREF Question 2.
f
Total score of GDS (short version). GDS Score 5 or above suggests depression, Score 10 or above is always indicative of depression.
g
Total score of PSQI. A total score of 5 or greater is indicative of poor sleep quality.

participants reported a significant reduction of GDS-15
score after receiving the treatment (P = 0.002). The mean
score of GDS-15 changed by −0.99 (95% CI, −1.60 to
−0.38) from 4.33 to 3.34, with an effect size of 0.288. The
number of depressed participants (GDS Score 8 or above)
was reduced from 15 (19%) to 8 (10%) after treatment.
PSQI: The PSQI mean score also dropped significantly
(P < 0.001) after treatment indicating that the sleep quality
of the participants had improved. The post-treatment PSQI
mean score changed by −1.81 (95% CI, −2.52 to −1.10),
from 9.75 to 7.94, with an effect size of 0.415. Detailed
analysis of the seven components of the PSQI scores
denoted that the positive change was mainly driven by the
participants’ improvement in daytime dysfunction, subject-
ive sleep quality and reduction of sleep disturbances.
Discussion
The results of our study indicated that applying a structured
acupressure protocol to the face, head and shoulder area of
the participants regularly for 12 weeks, 3–4 times a week,
did improve their physical health. It also had a positive
effect on their psychological well-being and social relationship.
The biggest improvement was found in the physical health
domain. The biggest single item change in WHOQOL-
BREF was also in the physical domain (Q3): the participants

961
C. W. C. Chan et al.

Table 3. Changes across study time points for 79 participants from both treatment and waitlist control groups who com-
pleted 80% of the interventiona—A per-protocol analysis.
Variable Mean ± standard Changes comparing to Within-group effect size P-valued
deviationb pre-treatment (95% CI) (Cohen’s d)c
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
WHOQOL-BREF (HK)
Self-reported overall QOL (Scores 1–5)e
Pre-treatment 3.48 ± 0.80
Post-treatment 3.62 ± 0.71 0.14 (−0.05 to 0.32) 0.185 0.138
3 months post-treatment 3.62 ± 0.71 0.14 (−0.04 to 0.33) 0.185 0.132
Self-reported General Health (Scores 1–5)f
Pre-treatment 3.06 ± 0.98
Post-treatment 3.27 ± 0.98 0.20 (−0.04 to 0.45) 0.214 0.101

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019

3 months post-treatment 3.30 ± 0.98 0.23 (−0.02 to 0.48) 0.245 0.065
Physical domain (Scores 0–100)
Pre-treatment 55.99 ± 16.35
Post-treatment 64.38 ± 15.64 8.39 (5.41 to 11.38) 0.524 <0.001*
3 months post-treatment 62.20 ± 15.38 6.22 (3.00 to 9.43) 0.391 <0.001*
Psychological domain (Scores 0–100)
Pre-treatment 64.56 ± 14.04
Post-treatment 66.11 ± 13.15 1.56 (−1.57 to 4.68) 0.114 0.325
3 months post-treatment 65.60 ± 8.44 0.96 (−2.31 to 4.24) 0.09 0.562
Social domain (Scores 0–100)
Pre-treatment 62.24 ± 11.82
Post-treatment 65.09 ± 10.58 2.85 (0.23 to 5.45) 0.254 0.032*
3 months post-treatment 63.83 ± 9.33 1.59 (−1.32 to 4.41) 0.149 0.268
Environment domain (Scores 0–100)
Pre-treatment 67.77 ± 11.82
Post-treatment 70.53 ± 12.27 2.76 (0.12 to 5.40) 0.229 0.041*
3 months post-treatment 69.47 ± 7.82 1.70 (−1.04 to 4.44) 0.170 0.222
Suffered from daily pain
Pre-treatment 47%
Post- treatment 32% −15% (−26% to −5%) 0.293 0.006*
3 months post-treatment 41% −6% (−19% to −8%) 0.124 0.397
Pain Intensity (Scores 0–10)
Pre-treatment 4.63 ± 3.02
Post-treatment 3.28 ± 2.84 −1.35 (−1.95 to −0.76) −0.461 <0.001*
3 months post-treatment 3.84 ± 2.84 −0.80 (−1.44 to −0.15) −0.269 0.016*
GDS (Scores 0–15)g
Pre-treatment 4.33 ± 3.73
Post-treatment 3.34 ± 3.11 −0.99 (−1.60 to −0.38) −0.288 0.002*
3 months post-treatment 3.35 ± 3.20 −0.98 (−1.64 to −0.31) −0.282 0.005*
PSQI (Scores 0–21)h
Pre-treatment 9.75 ± 4.53
Post-treatment 7.94 ± 4.18 −1.81 (−2.52 to −1.10) −0.415 <0.001*
3 months post-treatment 9.09 ± 4.44 −0.66 (−1.51 to −0.19) −0.147 0.125

*P < 0.05.
a
The 79 elderly participants’ participation rate on the treatment sessions delivered by the registered Chinese Medicine Practitioner is 84%, and their caregivers
reported a 90% compliance to the treatment schedule.
b
The estimated mean from linear mixed-effects model, SD converted from Standard Error generated from the same model.
c
Cohen’s d of 0.2 represents a small effect size, 0.5 a medium effect size and 0.8 a large effect size. Cohen’s d is generated by effect size calculator (https://www.
psychometrica.de/effect_size.html). For continuous variable, means and standard deviations converted from standard error generated by the mixed model were
used as input. For binary data, estimated number of success and failure generated by the mixed model were used as input.
d
P-value indicates comparison of each follow-up time-point to baseline, based on generalised mixed-effects model generated by SPSS procedure:
GENLINMIXED with time as a fixed effect. For continuous variables, an IDENTITY link is used; for binary variable, a LOGIT link is used.
e
Score of WHOQOL-BREF Question 1.
f
Score of WHOQOL-BREF Question 2.
g
Total score of GDS (short version). GDS Score 5 or above suggests depression, Score 10 or above is always indicative of depression.
h
Total score of PSQI. A total score of 5 or greater is indicative of poor sleep quality.

found that physical pain-induced constraints was signifi-
cantly reduced by the intervention. The participants also
experienced a significantly higher satisfaction in their ability
to perform daily living activities (Q17). Ability to do essen-
tial tasks and to maintain their independence proved to be
highly correlated to the QOL of the older people in prior
researches. Reduction of pain also allowed the older people
to be more active, and willing to participate in the commu-
nity, which may further enhance their QOL in the social
domain.

962
 Acupressure for frail older people in community dwellings
Significant reduction of the GDS score indicated that
the acupressure protocol had also made a positive impact
on the psychological well-being of the participants. In the
program evaluation, participants revealed that they felt hap-
pier and were in a better mood after the intervention. As
the psychological well-being of the older people is often
compromised by disability and/or weakened social support
systems [20], improvements in the physical health and social
relationship domains might also have contributed to the
reduction of GDS scores.
In our study cohort, a high percentage of participants
(79% of the 101 participants) suffered from poor sleep quality
before treatment (PSQI score ≥ 5). In a recent research con-
ducted on 1,086 healthy community residents aged 60 and
above in Shanghai China, the prevalence of poor sleep quality
(PSQI score ≥ 5) was found to be 41.5% [21]. The higher
prevalence of poor sleep quality in our study cohort indicated
that frailty and old age could be the risk factors of poor sleep
quality. Poor sleep quality had a negative impact on both the
physical health and psychological well-being of the older people
[22]. Despite the high prevalence of sleep problems among the
older adults, it is normally undertreated, and non-pharmacologic
interventions are underused [23]. Our study’s findings have
presented a viable non-phamacological treatment to improve
the sleep quality of the frail older people and possibly other
older people who suffered from the same syndomes.
Owing to the limitations of our study design, we can
only report the 3 months post-treatment data for the 79 par-
ticipants who completed the treatment with 80% compliance,
and are unable to provide 3 months post-treatment compari-
son between the treatment and control groups.
In our 3 months follow-up, the post-treatment improve-
ments in some domains subsided, but the physical domain
score of WHOQOL_BREF, alleviation of pain and reduc-
tion of GDS score remained significant. We recommend
the acupressure protocol to be administered on the frail
older people on a regular and continual basis. To ensure its
sustainability, the protocol was designed as a caregiver
administered treatment, which could be carried out at home
or any community settings. Another key limitation to this
study is that we could not blind the participants. The treat-
ment group knows that they are receiving treatment with
attentive care. The control group activity could account for
the additional social interaction time and attention but may
not be able to totally account for the positive effect gener-
ated by the participant’s belief in the treatment. The self-
reported outcome measure could also potentially be a
source of bias.
Conclusion
The study outcomes indicate that the acupressure protocol
used in this study is a safe and effective non-invasive treat-
ment that can benefit the frail older people living in com-
munity dwellings. The between-group differences in change
after 12 weeks indicating that the intervention has a positive
impact on the frail older people’s QOL, especially in the
physical domain. It also serves to reduce pain and improve
sleep quality. With proper training, informal caregivers can
apply this acupressure protocol regularly on the older people.
We recommend this acupressure protocol to be promoted in
the community as a caregiver administered treatment for the
frail older people. Programs should be developed in collabor-
ation with Chinese medicine practitioners to train, and pro-
vide support to the practicing caregivers in order to ensure
the sustainability of the intervention.
Key points
Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019
• Acupressure is a non-pharmacological, non-invasive Trad-
itional Chinese Medicine (TCM) treatment which can
improve the quality of life in frail older population.
• Acupressure could improve frail older adults’ physical
condition and sleeping quality and reduce physical pain.
• TCM-principle guided acupressure intervention used in
this study could be promoted as a caregiver administered
intervention.
Supplementary data
Supplementary data are available at Age and Ageing online.
Conflicts of interests
None declared.
Funding
Social Welfare Development fund, Social Welfare Department
of Hong Kong, China (through Yan Chai Hospital Social
Welfare Department).
References
1. Buckinx F, Rolland Y, Reginster JY, Ricour C, Petermans J,
Bruyère O. Burden of frailty in the elderly population: per-
spectives for a public health challenge. Arch Public Health
2015; 73: 19.
2. Fried LP, Darer J, Walston J. Frailty. In: Cassel C, Leipzig R,
Cohen H, Larson E, Meier D (eds). Geriatric Medicine: An
Evidence-Based Approach, 4th edition. New York: Springer-
Verlag, 2003: pp. 1067–74.
3. Lee EJ, Frazier SK. The efficacy of acupressure for symptom
management: a systematic review. J Pain Symptom Manage
2011; 42: 589–603.
4. Cao H, Pan X, Li H, Liu J. Acupuncture for treatment of
insomnia: a systematic review of randomized controlled trials.
J Alternat Complement Med 2009; 15: 1171–86.
5. Yeung WF, Chung KF, Poon MM et al. Acupressure, reflexology,
and auricular acupressure for insomnia: a systematic review of
randomized controlled trials. Sleep Med 2012; 13: 971–84.
6. Sarris J, Byrne GJ. A systematic review of insomnia and
complementary medicine. Sleep Med Rev 2011; 15: 99–106.
963
J. Johansson et al.

7. Yeung WF, Chung KF, Leung YK, Zhang SP, Law AC.
Traditional needle acupuncture treatment for insomnia: a sys-
tematic review of randomized controlled trials. Sleep Med
2009; 10: 694–704.
8. Chen YW, Wang HH. The effectiveness of acupressure on
relieving pain: a systematic review. Pain Manag Nurs 2014;
15: 539–50.
9. Gobbens RJ, van Assen MA, Luijkx KG, Wijnen-Sponselee
MT, Schols JM. The Tilburg frailty indicator: psychometric
properties. J Am Med Dir Assoc 2010; 11: 344–55.
10. Chu LW, Pei CK, Ho MH, Chan PT. Validation of the abbre-
viated mental test (Hong Kong version) in the elderly medical
patient. Hong Kong Med J 1995; 1: 207–11.
16. Woo J, Ho SC, Lau J et al. The prevalence of depressive
symptoms and predisposing factors in an elderly Chinese
population. Acta Psychiatr Scand 1994; 89: 8–13.
17. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ.
The Pittsburgh Sleep Quality Index: a new instrument for psy-
chiatric practice and research. Psychiatry Res 1989; 28: 193–213.
18. Tsai PS, Wang SY, Wang MY et al. Psychometric evaluation
of the Chinese version of the Pittsburgh Sleep Quality Index
(CPSQI) in primary insomnia and control subjects. Qual Life
Res 2005; 14: 1943–52.
19. Lenhard, W & Lenhard, A (2016). Calculation of Effect Sizes.
available: https://www.psychometrica.de/effect_size.html.
Bibergau (Germany): Psychometrica.

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/957/3787760 by guest on 16 March 2019

11. Harper A. Development of the World Health Organisation
WHOQOL-BREF quality of life assessment. Psychol Med
1998; 28: 551–8.
12. Hwang HF, Liang WM, Chiu YN, Lin MR. Suitability of the
WHOQOL-BREF for community-dwelling older people in
Taiwan. Age Ageing 2003; 32: 593–600.
13. Leung KF, Wong WW, Tay MS, Chu MM, Ng SS. Development
and validation of the interview version of the Hong Kong
Chinese WHOQOL-BREF. Quality Life Res 2005; 14: 1413–9.
14. Sheikh JI, Yesavage JA. Geriatric Depression Scale (GDS):
Recent evidence and development of a shorter version. Clin
Geront 1986; 5: 165–73. NY: The Haworth Press.
15. Lam CW, Boey KW. The psychological well-being of the
Chinese elderly living in old urban areas of Hong Kong: a
social perspective. Ageing Ment Health 2005; 9: 162–6.
20. Canadian Mental Health Association Newfoundland and
Labrador. The Mental Health Needs of an Ageing Population.
[cited 2016Feb25]. Retrieved from: http://www.cmhanl.ca/
pdf/Healthy%20Ageing.pdf.
21. Luo J, Zhu G, Zhao Q et al. Prevalence and risk factors of
poor sleep quality among Chinese elderly in an urban com-
munity: results from the Shanghai ageing study. PloS one
2013; 8: e81261.
22. Roth T. Insomnia: definition, prevalence, etiology, and conse-
quences. J Clin Sleep Med 2007; 3: S7.
23. Kamel NS, Gammack JK. Insomnia in the elderly: cause,
approach, and treatment. Am J Med 2006; 119: 463–9.
Received 3 October 2016; editorial decision 22 March
2017

Age and Ageing 2017; 46: 964–970 © The Author 2017. Published by Oxford University Press on behalf of the British Geriatrics Society.
doi: 10.1093/ageing/afx083 All rights reserved. For permissions, please email: journals.permissions@oup.com
Published electronically 22 May 2017

Increased postural sway during quiet stance

as a risk factor for prospective falls in
community-dwelling elderly individuals
JONAS JOHANSSON1,2, ANNA NORDSTRÖM1, YNGVE GUSTAFSON2, GÖRAN WESTLING3, PETER NORDSTRÖM2
1
Department of Public Health and Clinical Medicine, Occupational and Environmental Medicine, Umeå University, Umeå, Sweden
2
Department of Community Medicine and Rehabilitation, Geriatric Medicine, Umeå University, Umeå, Sweden
3
Department of Integrative Medical Biology, Physiology Section, Umeå University, Umeå, Sweden
Address correspondence to: P. Nordström. Tel: +46 90 786 0000. Fax: +46 90 133 607. Email: peter.nordstrom@umu.se

Abstract

Objective: fall-related injuries constitute major health risks in older individuals, and these risks are projected to increase in
parallel with increasing human longevity. Impaired postural stability is a potential risk factor related to falls, although the evi-
dence is inconclusive, partly due to the lack of prospective studies. This study aimed to investigate how objective measures
of postural sway predict incident falls.
Design, setting and participants: this prospectively observational study included 1,877 community-dwelling individuals
aged 70 years who participated in the Healthy Ageing Initiative between June 2012 and December 2015.

964