Ellegaard et al.

Arthritis Research & Therapy (2019) 21:158

https://doi.org/10.1186/s13075-019-1924-9

RESEARCH ARTICLE Open Access

Hand exercise for women with rheumatoid

arthritis and decreased hand function: an
exploratory randomized controlled trial
Karen Ellegaard1*, Cecilie von Bülow1,2, Alice Røpke3, Cecilie Bartholdy1,4, Inge Skovby Hansen4,
Signe Rifbjerg-Madsen1, Marius Henriksen1,4 and Eva Ejlersen Wæhrens1,2

Abstract
Background: People with hand-related rheumatoid arthritis (RA) experience problems performing activities of daily
living (ADL). Compensatory strategies to improve ADL ability have shown effective. Similarly, hand exercise has shown
effect on pain, grip strength, and self-reported ability. A combination has shown positive effects based on self-report,
but self-report and observation provide distinct information about ADL. The purpose of this study was to examine
whether hand exercise as add on to compensatory intervention (CIP) will improve observed ADL ability in RA.
Methods: Women (n = 55) with hand-related RA were randomized to CIPEXERCISE (intervention) or CIP only (control). CIP
is focused on joint protection, assistive devices, and alternative ways of performing AD. The hand-exercise program
addressed range of motion and muscle strength.
Primary outcome was change in observed ADL motor ability measured by the Assessment of Motor and Process Skills
(AMPS). Baseline measures were repeated after 8 weeks.
Results: Improvements in ADL motor ability in CIPEXERCISE (mean change = 0.24 logits; 95% CI = 0.09 to 0.39) and
CIPCONTROL (mean change =0.20 logits; 95% CI = 0.05 to 0.35) were statistically significant, with no differences between
groups (mean difference = 0.04 logits; 95% CI = − 0.16 to 0.25). Thirteen (46.4%) participants in the CIPEXERCISE and 12
(44.4%) in the CIPCONTROL obtained clinically relevant improvements (≥ 0.30 logits) in ADL motor ability; this group
difference was not significant (z = 0.15; p = 0.88).
Conclusion: Adding hand exercise to a compensatory intervention did not yield additional benefits in women with
hand-related RA.
The study was approved by the ethics committee 14th of April 2014 (H-3-2014-025) and registered at ClinicalTrials.gov
16th of May 2014 (NCT02140866).
Keywords: Rheumatoid arthritis, Hand function, ADL performance, Exercise program, Observational based assessment

Introduction Hand-exercise programs may have positive effects on

Rheumatoid arthritis (RA) is a chronic autoimmune dis- hand function in RA [5–9]. A randomized trial including
ease characterized by inflammation of synovium [1]. In patients with RA-related hand problems showed positive
90% of RA patients, joints of the hand are affected result- effects on hand function after hand exercise added to a
ing in problems performing activities of daily living (ADL) 1.5-h instruction in joint protection, without negative ef-
[2, 3]. A study by Thyberg et al. indicates that low grip fect on disease activity [8–10].
strength may play a role in decreased ADL ability [4]. Compensatory programs based on individualized joint
protection and education in coping strategies to improve
performance of ADL tasks have shown to be effective
* Correspondence: [email protected]
[11–13]. Hammond et al. found that patients with RA,
1
The Parker Institute, Copenhagen University Hospital Bispebjerg and diagnosed within < 5 years, attending an educational-
Frederiksberg, Nordre Fasanvej 57, DK-2000 Copenhagen F, Denmark behavioral joint protection program, maintained ADL
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 2 of 9

ability after 4 years [11, 12]. Masiero et al. showed that are as follows: significant osteoarthritis of the hand
patients with RA presented with less pain and disability (assessed by the rheumatologist); hand surgery within
8 months after an educational-behavioral joint protec- 6 months; other pain condition involving muscles and/
tion program [13]. or joints; prednisolone therapy; alternative treatments
Clinical guidelines recommend that patients with RA during study period; change in medical treatment during
receive hand-exercise programs focusing on increasing study period; inability to understand Danish; and finally
muscle strength and movements and programs on com- any other contradictions for participating assessed by
pensatory strategies to overcome difficulties in ADL task the rheumatologist. Participants were asked not to par-
performance related to hand impairments [14]. ticipate in occupational or physical therapy interventions
However, existing evidence of the effectiveness of hand elsewhere during the study period, aquatic physical ther-
exercise on functional ability is based on self-reported apy was allowed.
data, typically questionnaires. While self-report repre-
sents the patient’s perspective, it is documented that Ethics approval and consent to participate
measures of self-reported ADL ability have low to mod- Participants gave written informed consent. The study
erate relationship to observation-based ADL ability mea- was approved by the ethics committee of the Capital Re-
sures [3, 15]. Thus, self-reported and observation-based gion of Denmark (H-3-2014-025) and registered at Clini-
measures provide distinct but complementary informa- calTrials.gov (NCT02140866). The study was conducted
tion [3, 15]. Observation-based evaluations of ADL abil- in accordance with standards of the responsible commit-
ity are considered to provide more neutral measures tee on human experimentation and the Declaration of
than evaluations based on self-report, as observations Helsinki. If participants experienced increased disease
seem less influenced by psychosocial factors and pain activity during the study period, US and rheumatologist
[16]. Furthermore, observation-based evaluations have examinations were made to determine if modifications
shown to be more sensitive to change after intervention or withdrawal were needed.
among patients with chronic pain [17].
It remains to be tested whether hand-exercise therapy Study design and randomization
as an add on to a compensatory intervention program The investigator-initiated study was designed as an RCT
(CIP) will improve the observable performance of ADL with parallel groups. Participants entered an 8-week pro-
tasks requiring hand function in patients with RA. gram and were randomly assigned to either intervention
Moreover, as suppression of disease activity is essential group (hand-exercise therapy and CIP, CIPEXERCISE) or con-
to avoid progression of joints destruction [18, 19], it is trol group (CIP alone, CIPCONTROL) with a 1:1 equal alloca-
relevant to explore changes in disease activity after tion ratio utilizing a concealment process. Randomization
therapeutic interventions. was made using sealed envelopes and carried out by the pro-
It was hypothesized that hand-exercise therapy as an ject secretary. The outcome assessors and data analysts were
add on to CIP in patients with RA-related hand impair- kept blinded to the allocation, and participants were
ments would result in larger improvements in observed instructed not to convey their group allocation. The persons
ADL ability as compared to CIP alone, without increas- performing US examination and evaluation were blinded to
ing disease activity. results of clinical examination and group allocation.

Patients and methods Intervention programs

Participants The CIP consisted of an introduction to compensatory
Participants were females with RA reporting ADL task strategies including joint protection (JP), assistive devices,
performance problems involving the hands, recruited and alternative methods of performing ADL [21, 22]. The
from May 2014 to January 2016 through rheumatologists program was client-centered focusing on improving the
at outpatient clinics in Copenhagen and announcements ability to perform those ADL tasks that the single partici-
in daily press. Potential participants were given further in- pant identified as purposeful to her life. It consisted of
formation and pre-screened for eligibility via telephone. If four 1-h sessions during an 8-week period. One occupa-
inclined to participate, they were examined by a rheuma- tional therapist (OT) (ISH) with > 10 years of experience
tologist to determine if they fulfilled inclusion criteria. performed all sessions. The first session aimed to identify
Inclusion criteria are as follows: females aged > 18 the participant’s individual resources and problems in
years; diagnosed with RA (ACR/EULAR 2010 Criteria) ADL task performance, goal setting focused on ADL task
[20]; involvement of minimum of one tender wrist, performance, and introduction to assistive devices. As
MCP, or PIP joint; stable medication 3 months prior to homework, assistive devices were applied at home and the
participation, and self-reported decreased ability to per- utility evaluated. In the second session, the OT presented
form ADL tasks involving the hands. Exclusion criteria JP principles and discussed these in relation to the
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 3 of 9

participant’s ADL task problems. She supervised the par- ADL-Q), self-reported disability (Stanford Health Assess-
ticipant to integrate JP principles and alternative methods ment Questionnaire Disability Index, HAQ-DI), overall
of doing in ADL tasks. In the third session, the OT disease activity (DAS28), grip strength, and pain. Explora-
followed up on the use of JP principles and assistive de- tory outcome was disease activity assessed by US.
vices at home and together they identified any additional
ADL task, representing a challenge, and performed this AMPS
under supervision. Optionally, additional assistive devices The AMPS is a standardized observation-based tool
were handed out. As homework, the participant practiced used to measure a person’s observed ADL task perform-
JP principles and assistive devices in the chosen task. ance [27, 28]. The person being evaluated chooses and
Fourth session, the OT followed up on homework, evalu- performs at least two standardized ADL tasks of
ated goals, and introduced how to apply for assistive de- relevance and appropriate challenge. During AMPS
vices and housing accessibility solutions in the home evaluation, two domains are evaluated, i.e., ADL motor
municipality. This fourth session, if relevant, was done as ability (the amount of effort, fatigue and/or clumsiness)
a telephone meeting. and ADL process ability (the degree of disorganization,
The hand-exercise program lasted for 8 weeks as a inappropriate use of time, space, objects and ability to
strength increase is possible within this time frame and adapt actions). The 16 ADL motor and 20 ADL process
was designed based on recent research [6, 8, 23–25]. skills are evaluated in terms of ease, efficiency, safety,
First, the exercise intervention was performed four times and independence using a four-point ordinal scale. The
per week with one session supervised by a physical ther- available AMPS software [29], based on a many-faceted
apist (PT), and the other three were home based. After Rasch measurement model, makes it possible to convert
14 days. One PT (CB) with 3 years of experience per- ordinal raw scores into overall linear ADL motor ability
formed the individual exercise sessions. Home-based ex- measures and overall linear ADL process ability mea-
ercise sessions were recorded in a diary. During the sures adjusted for task challenge, skill item difficulty,
period, the exercise intervention was increased to once and rater severity. Measures are expressed in logits (log-
daily and the load was increased. The program consisted odds probability units) [28]. The overall ADL motor
of three parts: (1) warm-up/mobility (10 min), (2) muscle ability measure indicates how much effort or fatigue the
strength training (20 min), and (3) cool-down (5 min). person demonstrated, and the overall ADL process abil-
The warm-up was performed to prepare the joints for ity measure indicates how efficient the person was ob-
the muscle strength training and to improve flexibility; served to be during the ADL task performance.
the muscle strength training was designed to ensure that Additionally, both ADL ability measures reflect safety
relevant muscle was targeted within a period of 20 min. and independence in ADL task performance. ADL ability
Resistance was supplied by exercise bands and Thera- measures above the 2.0 logit cutoff on the ADL motor
putty. The amount of resistance was based on the weak- scale and above the 1.0 logit cutoff on the ADL process
est hand. The resistance intensity was set according to scale indicate effortless, efficient, safe, and independent
the participant’s self-reported experience of load using ADL task performance in everyday life. In contrast, ADL
the Borg Scale [26]. This load was chosen to minimize motor ability measures below the 2.0 logits cutoff indi-
the risk of flare-up symptoms due to overload and to cate increased effort or fatigue during task performance.
allow for progression. For cool-down, some of the Moreover, ADL ability measures below the 1.50 ADL
warm-up exercises were receded. All participants re- motor cutoff and/or below the 1.00 ADL process cutoff
ceived a detailed illustrated description of the exercise indicate a need for minimal assistance for community
program (Additional file 2). In case of any flare-up in living. Finally, according to the AMPS manual a differ-
symptoms, the participant was set to only conduct the ence of > 0.30 logits on the AMPS ADL motor and/or
warm-up and cool-down part. Post exercise soreness ADL process scale defines a clinically relevant change
and temporary fatigue was tolerated. Detailed description [28]. Studies support that the AMPS ability measures are
of the exercise program is presented in Additional file 3. If reliable and valid in RA patients [3].
a participant failed to meet at an appointment, a phone
call was made to maintain fidelity. ADL-Q
Self-reported ADL ability was assessed using the ADL-
Outcomes Q, a standardized instrument with 47 ADL tasks devel-
Primary outcome was observed, ADL motor ability mea- oped to measure a person’s perceived quality of ADL
sured by Assessment of Motor and Process Skills performance [3, 30]. The person marks the quality of the
(AMPS). ADL task performance using seven response categories
Secondary outcomes were observed, ADL process ability reflecting efficiency, effort/fatigue, safety, and independ-
(AMPS), self-reported ADL ability (ADL-Questionnaire, ence. Rasch methods are applied [30]. Studies support
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 4 of 9

that the ADL-Q can be used to generate valid measures two-sided significance level of 5% (p ≤ 0.05), assuming a
of self-reported quality of ADL task performance among common standard deviation (SD) of 0.36 logits, a sample
RA patients [3]. size of n = 32 participants per group was required to ob-
tain a power of at least 90% to detect a group mean dif-
HAQ-DI ference of 0.3 logits. It was decided to include n = 45
The HAQ-DI is developed to assess disability in RA [21]. participants in each group.
The questionnaire consists of 20 questions primarily con-
cerning ADL tasks, a Danish version was used [31]. Statistical analysis
Data analyses were carried out according to a pre-
DAS28 established statistical analysis plan (SAP); all analyses
Overall disease activity was assessed using the DAS28 were done applying SAS (v. 9.4, SAS Institute Inc., Cary,
which is based on count of 28 joints for swelling and NC, USA). Descriptive statistics and tests are reported
tenderness, C-reactive protein level in the blood and the in accordance to the “Enhancing the QUAlity and
patient’s self-reported impact of disease on a visual ana- Transparency Of health Research” (EQUATOR) network
log scale (VAS Global health). The DAS28 score range [38, 39]. To evaluate the empirical distributions of the
from 0 to 9.4 [32]. continuous outcomes, visual inspection was applied to
suggest whether the assumption of normality was rea-
Grip strength sonable. The PROCUNIVARIATE statement was used
The maximal grip strength was measured in kilogram for summarizing the data (descriptive statistics).
using a digital hand Dynamometer (North Coast Medical Intention-to-treat (ITT) analyses were made, i.e., ana-
Inc.). The grip strength was measured three times in lyzing participant outcomes according to the group to
both hands. The maximal force performed in each hand which they were allocated, even if participants did not
was used in the analysis. receive allocated intervention. The ITT principle was
done by replacing missing data with the value obtained
Pain at baseline.
Hand pain during activity and in rest was measured in At week 8, the CIPEXERCISE group was compared with
both hands on a visual analog scale (VAS), where zero the CIPCONTROL group using general linear model (ana-
was no pain and 100 was maximal pain. lyses of covariance; ANCOVA) for mean changes from
baseline and t tests for comparison of least squares
Ultrasound examination means between groups. The model included change as
Synovial hypertrophy and increased synovial perfusion the dependent variable (Δ), with treatment group as a
are indications of disease activity assessed by US [33]. main effect and the baseline score as an additional
Gray scale US is used to examine synovial hypertrophy covariate. Results were expressed as the difference between
seen as hypo-echoic structure between the tendon/ the group means and 95% CI with the associated p values,
muscle and the bone [1]. Doppler US added to the gray based on the general linear model.
scale image register movement of the blood as an indica- For sensitivity purposes, the analyses were repeated
tion of increased synovial perfusion. US has shown to with further adjustment for disease duration and hand
correlate with measures of disease activity in RA [19, 34, pain at baseline as there were group imbalances in these
35]. US examination in RA has shown to display sub variables at baseline (Additional file 1: Table S1).
clinical disease activity leading to disease progression on The proportion of participants responding to therapy
X-ray [36]. The wrist and MCP 2–5 were examined both (≥ 0.30 logits on the AMPS ADL motor scale) was ana-
dorsal and palmar. The wrist was scanned in four dorsal lyzed using z test to evaluate the difference in the num-
and one volar position and the MCP joints in three dor- ber of responders between groups.
sal and one volar position. Both synovial hypertrophy
and Doppler were evaluated using a validated scoring Results
system for RA [37]. One score for synovial hypertrophy, Fifty-five were randomized to CIPEXERCISE (n = 28) or
one score for synovial perfusion (Doppler), and one sum CIPCONTROL (n = 27); 22 and 25 participants, respect-
score were calculated. ively, completed the trial (Fig. 1). Reason for drop out in
CIPCONTROL was change in medication. In CIPEXERCISE,
Sample size one changed medication, three had flare-up in other dis-
Sample size calculation was based on previous data on eases, one died of another disease, and one found the
AMPS ADL motor ability in women with RA [3]. For a intervention too time consuming. Due to loss of the
two sample pooled t-test of a normal mean difference AMPS ADL motor ability measure at baseline for one
with pooled variances (equal variances assumed) and a participant in CIPEXERCISE, the ITT population consisted
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 5 of 9

Fig. 1 CONSORT flow diagram of the study

of n = 27 in each group. The ITT population’s mean age with no statistically significant difference between
was 63.7 (SD = 13.0) years, and mean disease duration groups (ADL motor mean difference = 0.04 logits; 95%
was 12.6 (SD = 11.1) years. Mean baseline tender and CI = − 0.16 to 0.25). A significant increase in grip
swollen joint count (28 joints) was 5.0 (SD = 4.8) and 1.4 strength in the right hand was seen in the CIPEXERCISE
(SD = 1.7), respectively; the mean hand pain in activity group and ESR increased significantly in the CIPCON-
was 42.7 (SD = 26.1) (right) and 36.2 (SD = 27.6) (left) TROL group (Table 2).
mm VAS. Mean baseline AMPS ADL motor ability The sensitivity analyses confirmed the primary ana-
measure was 1.4 (SD = 0.5) logits. Mean baseline HAQ lyses and supported further the tendencies towards
score was 1.1 (SD = 0.6). For other baseline characteris- group differences in DAS28, HAQ-DI, and ultrasound
tics, see Table 1. total score (Additional file 1: Table S1).
The CIPEXERCISE group attended a mean of 2.4 (SD = The responder analysis revealed that 13 (46.4%) partic-
1.2) CIP sessions, whereas the participants in the CIPCON- ipants in the CIPEXERCISE and 12 (44.4%) in the CIPCON-
TROL attended 2.7 (SD = 1.0) CIP sessions. The average TROL obtained clinically relevant improvements (≥ 0.30)
number of hand-exercise sessions was 20.2 (SD = 10.3). in ADL motor ability, and this difference was not signifi-
Exercise diary was received from 24 of the 27 participants. cant (z = 0.15; p = 0.88).
The hand-exercise program was delivered as intended (see
Additional file 3). Discussion
No significant mean differences in changes from base- Based on the results of this study, it was not possible to
line were seen between groups in any of the outcomes confirm the hypothesis that hand-exercise therapy as
(Table 2). Still, tendencies towards differences between add on to a compensatory intervention program in pa-
groups in changes in DAS28 score, CRP, and ESR were tients with decreased ADL ability, following RA-related
seen. Improvements in primary outcome, observed ADL hand impairments, would result in larger improvements
motor ability, were seen in both groups, in the CIPEXER- in observed ADL ability as compared to CIP alone.
CISE (ADL motor mean change = 0.24 logits; 95% CI = While no statistically significant differences in changes
0.09 to 0.39) and CIPCONTROL (ADL motor mean in primary and secondary outcomes were seen between
change = 0.20 logits; 95% CI = 0.05 to 0.35), however groups, both groups obtained statistically significant
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 6 of 9

Table 1 Baseline characteristics for all randomized participants

CIPEXERCISE (n=27) CIPCONTROL (n=27) Total (54)
Mean (SD)/ Median (IQR) Mean (SD)/ Median (IQR) Mean (SD)/ Median (IQR)
Age (Years) 64.8 (13.5) 62.6 (12.0) 63.7 (12.8)
Weight (Kg) 69.4 (12.3) 70.0 (11.9) 70.2 (12.5)
Height (Cm) 167.6 (7.4) 166.2 (7.4) 166.9 (7.4)
Start symptoms (Years) 11.3 (3.5:22.5) 14.1 (7.9:18.2) 11.9 (4.7:20.3)
Disease duration (Years) 3.1 (0.4:11.6) 10.3 (7.3:14.6) 7.6 (1.6:13.5)
AMPS ADL motora (Logits) 1.34 (0.4) 1.38 (0.5) 1.36 (0.5)
AMPS ADL processa (Logits) 1.20 (0.2) 1.13 (0.2) 1.27 (0.2)
ADL-Questionnaire (Logits) 0.4 (0.1) 0.5 (0.5) 0.5 (0.3)
HAQ-DIb 1.1 (0.6) 1.1 (0.6) 1.1 (0.6)
Hand strength and pain
Maximal grip strength (Kg)
Right 17.7 (7.7) 18.7 (8.1) 18.2 (7.8)
Left 17.2 (6.3) 17.8 (5.7) 17.5 (6.0)
Hand pain: Activity (mm VAS)
Right 50 (25.0:62.0) 32 (19.5:68.5) 40.5 (21.25:63:75)
Left 50 (11.0:65.5) 26 (10.5:52.0) 30.5 (11.0:60.25)
Hand pain: Rest (mm VAS)
Right 29 (7.0:54.5) 19 (10.0:50.0) 22 (10.0:51.5)
Left 39 (10.0:50.0) 14 (5.5:25.5) 16.5 (6.25:42.5)
Medication: PainKillers (Number per day) 1.7 (1.2) 1.6 (1.0) 1.6 (1.1)
DAS28c 3.6 (1.1) 3.2 (1.1) 3.4 (1.1)
Tender joint count (28 joints) 5.8 (5.7) 4.1 (3.7) 5.0 (4.8)
Swollen joint count (28 joints) 1.5 (1.7) 1.3 (1.7) 1.4 (1.7)
d
CRP 5.3 (6.6) 5.4 (9.8) 5.3 (8.1)
Disease activity (mm) (VAS) 47.19 (26.86) 41.19 (24.93) 44.19 (25.85)
e
ESR 16.1 (11.3) 15.2 (12.7) 15.6 (11.9)
UltraSound (US)
US score: synovial hypertrophy (0-126) 16.5 (15.6) 20.4 (23.2) 18.5 (19.8)
Synovial perfusion (Doppler activity) (0-126) 5.6 (11.6) 6.7 (12.8) 6.2 (12.1)
US score: sum (0-252) 22.2 (25.9) 27.1 (35.2) 24.7 (30.9)
a
AMPS Assessment of Motor and Process Skills
b
HAQ-DI Stanford Health Assessment Questionnaire Disability Index
c
DAS28 Disease Activity Score 28
d
CRP C-reactive protein
e
ESR Erythrocyte Sedimentation Rate

improvements in the primary outcome, observed ADL that increasing strength and flexibility of the hands
motor ability. While the mean increases in ADL motor would translate into improved ADL ability. While this
ability in both groups were not clinically relevant, re- was not the case, participants in the CIPEXERCISE group
sponder analysis revealed that almost half of the partici- still obtained a significant increase in grip strength,
pants (46.3%) achieved a clinically relevant (i.e., > 0.3 which was not observed in the CIPCONTROL group. This
logits) increase in observed ADL ability. These results, suggests that hand exercise may improve strength, but
and the fact that the number of responders in each group that such improvements not necessarily translate into
was almost the same, suggest that a large percentage of improved ADL ability. This indication of a benefit of the
the participants benefitted from the CIP intervention. exercise program is further implied by a minor decrease
The CIPEXERCISE group received hand exercise as a re- in DAS28 score seen in the CIPEXERCISE group as com-
storative add on to the CIP intervention. It was assumed pared to the CIPCONTROL group. However, the change in
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 7 of 9

Table 2 Changes from Baseline in Primary and Secondary Outcomes Analyses based on the Intention-To-Treat-Population
CIPEXERCISE (n=27) CIPCONTROL (n=27 ) Group difference P-value
Mean change Mean change Mean change (95%CI)
(95% CI) (95% CI)
AMPSa ADL motor ability 0.24 (0.09 to 0.39) 0.20 (0.05 to 0.35) 0.04 (-0.16 to 0.25) 0.70
a
AMPS ADL process ability 0.05 (-0.05 to 0.14) 0.02 (-0.07 to 0.12) 0.02 (-0.11 to 0.16) 0.73
ADL-Questionnaire (ADL-Q) 0.09 (-0.01 to 0.20) 0.09 (-0.02 to 0.19) 0.01 (-0.15 to 0.16) 0.90
b
HAQ-DI -0.09 (-0.19 to 0.01) 0.01 (-0.09 to 0.11) -0.1 (-0.24 to 0.04) 0.16
Hand strength and pain
Max grip strength (Kg)
Right 1.43 (0.40 to 2.45) 0.18 (-0.86 to 1.23) 1.24 (-0.22 to 2.71) 0.10
Left 1.00 (-0.47 to 2.47) 0.36 (-0.71 to 3.44) 1.36 (-0.71 to 3.44) 0.20
Hand pain, Activity (mm VAS
Right -1.10 (-7.90 to 5.78) 0.57 (-6.27 to 7.41) -1.63 (-11.30 to 8.04) 0.74
Left -5.26 (-12.80 to 2.28) -0.78 (-8.31 to 6.76) -4.48 (-15.28 to 6.30) 0.41
Hand pain, Rest (mm VAS)
Right -1.43 (-8.63 to 5.78) 2.87 (-4.33 to 10.07) -4.30 (-14.49 to 5.00) 0.40
Left -1.67 (-9.71 to 6.36) 4.41-(-3.62 to 12.45) -6.08 (-17.80 to 5.62) 0.30
PainKillers (Number per day) 0.01 (-0.19 to 0.22) 0.01 (-0.22 to 0.19) 0.02 (-0.26 to 0.31) 0.87
c
DAS28 -0.17 (-0.49 to 0.15) 0.26 (-0.10 to 0.53) -0.39 (-0.84 to 0.07) 0.09
Tender joint count -0.51 (-1.83 to 0.82) 0.38 (-0.93 to 1.68) -0.88 (-2.75 to 0.99) 0.35
Swollen joint count 0.23 (-0.27 to 0.72) 0.15 (-0.33 to 0.63) 0.08 (-0.61 to 0.77) 0.82
CRPd -0.67 (-2.44-1.11) 1.76 (-0.12-3.63) -2.42 (-5.00-0.15) 0.06
Disease activity (VAS) -7.97 (-15.86—0.09) -0.70 (-7.93 to 6.53) -7.27 (-17.97 to 3.43) 0.18
ESRe -0.50 (-2.52 to 1.51) 2.30 (0.29 to 4.32) -2.80 (-5.65 to 0.05) 0.05
Ultra Sound (US)
US score: synovial hypertrophy 0.18 (-2.0 to 2.32) 1.83 (-0.22 to 3.89) -1.66 (-4.63 to 1.32) 0.27
Synovial perfusion (Doppler) -0.38 (-2.47 to 1.72) 1.01 (-0.99 to 3.03) -1.39 (-4.29 to 1.51) 0.34
UL score total -1.81 (-5.76 to 2.34) 2.56 (-1.23 to 6.36) -4.37 (-9.86 to 1.11) 0.12
a
AMPS Assessment of Motor and Process Skills
b
HAQ-DI Stanford Health Assessment Questionnaire Disability Index
c
DAS28 Disease Activity Score 28
d
CRP C-reactive protein
e
ESR Erythrocyte Sedimentation Rate
f
AntiCCP anti Cyclic Citrullinated Peptides

DAS28 was only minor and is not reaching the cutoff evaluations of ADL ability may be more sensitive to
for a clinically relevant change [40] and was not trans- measure changes following intervention than self-report.
lated into improved ADL ability. The hypothesis that participants receiving hand exer-
Despite statistically significant increases in observed cise would have no increase in disease activity was con-
ADL ability, similar changes in self-reported ADL ability firmed. Thus, no negative effect on disease activity after
were not seen across groups neither when using the exercise was seen either assessed locally by US examin-
diagnosis specific instrument HAQ-Di, nor the generic ation, by count of swollen and tender joints of the hand,
ADL-Q. These findings are in correspondence with the or globally as measured by inflammatory blood markers.
results reported in a study by Amris et al. [17]. In that In contrast, a significant increase in ESR was seen in the
study, a 2-week multi-component rehabilitation course CIPCONTROL group; however, ESR is a very slow reacting
resulted in improvements in observed ADL ability in pa- marker of disease activity and none of the other markers
tients with chronic widespread pain, but these improve- of disease activity showed the same tendency.
ments were not reflected in scores of self-reported In this randomized trial, the exercise program did not
functional abilities on standardized questionnaires. Thus, reduce hand-related pain. In other studies, investigating
our study supports the notion that observation-based the effect of hand-exercise programs in RA, there are
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 8 of 9

diverse findings regarding reduction in pain after exer- Funding

cise; in two studies, reductions in pain measured on a The study was supported by OAK foundation, Danish Physiotherapy
association, Etf., Denmark, Danish Rheumatism Association.
VAS scale was reported [5, 41], whereas two other
studies showed no pain reduction [6, 10]. The studies Declarations
reporting pain reduction did not evaluate functional The study was approved by the ethics committee capital region Denmark
(H-3-2014-025) and registered at ClinicalTrials.gov (NCT02140866).
ability [5, 41]. The two studies reporting no pain All participants gave written informed consent.
reduction evaluated and reported improvements in func-
tional ability (i.e., hand function, ADLs, pain, work Authors’ contributions
KE contributed in the design of study, ultrasound examination, and writing
performance, esthetics, and patient satisfaction with the manuscript. CvB contributed in performing AMPS and proof reading of
hand function) [8–10] and in evaluation of grip ability the manuscript. AR contributed in the design of the study and proof reading
[6]. These findings suggest that pain and functioning are of the manuscript. CB contributed in designing and performing the exercise
intervention and proof reading of the manuscript. IS contributed in
not necessarily closely linked factors, which is also seen performing the joint projecting education intervention. SRM contributed in
in a study investigating exercise therapy in patients with the doctor examination and proof reading of the manuscript. MH
impaired shoulder function [42]. Thus, the absence of contributed in the statistic analyses and proof reading of the manuscript.
EEW contributed in the design of the study and writing the manuscript. All
any significant effect of our hand-exercise intervention authors read and approved the final manuscript.
on ability cannot be explained by unchanged pain in the
hand after exercise. Authors’ information
Not applicable.
One limitation of the study was not reaching the
intended sample size. Still, the risk of overlooking a real Consent for publication
group difference in primary outcome is minor, as both Not applicable.
The datasets used and/or analyzed during the current study are available
groups obtained a statistically significant increase in from the corresponding author on reasonable request.
ADL motor ability with no indication of a group differ-
ence. Another limitation is that the disease duration was Competing interests
The authors declare that they have no competing interests.
longer in the CIPCONTROL compared to that in the CIP-
XERCISE despite the randomized design. However, no pro-
Publisher’s Note
nounced difference in any of the functional measures, Springer Nature remains neutral with regard to jurisdictional claims in
grip strength or disease activity score (DAS28) was seen published maps and institutional affiliations.
between the two groups strongly indicating the groups
Author details
were comparable. 1
The Parker Institute, Copenhagen University Hospital Bispebjerg and
Frederiksberg, Nordre Fasanvej 57, DK-2000 Copenhagen F, Denmark. 2The
Research Initiative for Activity Studies and Occupational Therapy, General
Practice, Department of Public Health, University of Southern Denmark,
Odense, Denmark. 3Metropolitan University College, Institute for
Conclusion Occupational Therapy and Physiotherapy, Copenhagen, Denmark.
In conclusion, participants in both CIPCONTROL and 4
Department of Physical and Occupational Therapy, Copenhagen University
CIPEXERCISE improved their ADL motor ability, but no sta- Hospital Bispebjerg and Frederiksberg, Copenhagen, Denmark.
tistically significant difference between groups was seen. Received: 9 January 2019 Accepted: 22 May 2019
Moreover, no differences between groups were seen in
secondary outcomes. Thus, based on our results, it cannot
References
be concluded that hand exercise as an add on to compen- 1. Terslev L, Torp-Pedersen S, Savnik A, von der Recke P, Qvistgaard E,
satory intervention further improves observed ADL ability Danneskiold-Samsøe B, Bliddal H. Doppler ultrasound and magnetic
in persons with RA-related impaired hand function. resonance imaging of synovial inflammation of the hand in rheumatoid
arthritis: a comparative study. Arthritis Rheum. 2003;48:2434–41.
2. Flemming A, Benn RT, Corbett M, Wood PH. Early rheumatoid disease. II.
Patterns of joint involvement. Ann Rheum Dis. 1976;35:361–4.
Additional file 3. Wæhrens EE, Bliddal H, Danneskiold-Samsoe B, Lund H, Fisher AG.
Differences between questionnaire- and interview-based measures of ADL
ability and the association to observed ADL ability in women with
Additional file 1: Table S1. Changes from baseline in primary and rheumatoid arthritis, knee osteoarthritis and fibromyalgia. Scand J
secondary outcomes analyses based on the intention-to-treat-population Rheumatol. 2012;41:95–102.
adjusted for baseline value, disease duration, and hand pain at baseline. 4. Thyberg I, Hass UAM, Nordenskiold U, Gerdle B, Skogh T. Activity limitation
(DOCX 20 kb) in rheumatoid arthritis correlates with reduced grip force regardless of sex:
Additional file 2: Patient handout exercise program. (PDF 1.48 kb) the Swedish TIRA project. Arthritis Rheum. 2005;53:886–96.
5. Rønningen A, Kjeken I. Effect of an intensive hand exercise programme in
Additional file 3: Detailed description of the hand-exercise programme.
patients with rheumatoid arthritis. Scand J Occup Ther. 2008;15:173–83.
(PDF 386 kb)
6. Brorsson S, Hilliges M, Sollerman C, Nilsdotter A. A six-week hand exercise
programme improves strength and hand function in patients with
rheumatoid arthritis. J Rehabil Med. 2009;41:338–42.
Acknowledgements 7. Wessel J. The effectiveness of hand exercises for persons with rheumatoid
Not applicable. arthritis: a systematic review. J Hand Ther. 2004;17:174–80.
Ellegaard et al. Arthritis Research & Therapy (2019) 21:158 Page 9 of 9

8. Adams J, Bridle C, Dosanjh S, Heine P, Lamb SE, Lord J, McConkey C, 26. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc.
Nichols V, Toye F, Underwood MR, Willams MA, Williamson EM. 1982;14:377–81.
Strengthening and stretching for rheumatoid arthritis of the hand SARAH : 27. Fisher AG. Assessment of motor and process skills: user manual. 7th ed. Fort
design of a randomised controlled trial of a hand and upper limb exercise Collins: Three Star Press; 2010.
intervention. BMC Musculoskelet Disord. 2012;13:230. 28. Fisher AG. Assessment of motor and process skills: development,
9. Williams MA, Williamson EM, Heine P, Nichols V, Glover M, Dritsaki M, standardization, and administration manual. 7th ed. Fort Collins: Three Star
Dritsaki M, Adams J, Dosanjh S, Underwood M, Rahman MCC, Lord J Lamb Press; 2010.
SE. Strengthening and stretching for rheumatoid arthritis of the hand 29. AMPS computer scoring software version 9.0 (computer program). Fort
SARAH . A randomised controlled trial and economic evaluation. Health Collins: Three Star Press; 2010.
Technol Assess. 2015;19:1–222. 30. Wæhrens EE. Measuring quality of occupational performance based on self-
10. Lamb SE, Williamson EM, Heine PJ, Adams J, Dosanjh S, Dritsaki M, Glover report and observation: development and validation of instruments to
MJ, Lord J, McConkey C, Nichols V, Rahman A, Underwood M, Williams MA. evaluate ADL task performance (dissertation) Department of Community
Exercises to improve function of the rheumatoid hand SARAH: a Medicine and Rehabilitation, Occupational Therapy, Umeå University,
randomised controlled trial. Lancet. 2015;385(9966):421–9. Sweden 2010.
11. Hammond A, Freeman K. The long term outcomes of a randomised 31. Thorsen H, Hansen TM, McKenna SP, Sorensen SF, Whalley D. Adaptation
controlled trail of an educational-behavioural joint protection programme into Danish of the Stanford health assessment questionnaire HAQ and the
for people with rheumatoid arthiritis. Clin Rehabil. 2004;18:520–8. rheumatoid arthritis quality of life scale RAQoL. Scand J Rheumatol. 2001;30:
12. Hammond A, Bryan J, Hardy A. Effects of a modular behavioural arthritis 103–9.
education programme: a pragmatic parallel group randomized controlled 32. Prevoo ML, van’t Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van
trail. Rheumatology (Oxford). 2008;47:1712–8. Riel PL. Modified disease activity scores that include twenty-eight-joint
13. Masiero S, Boniolo A, Wassermann L, Machiedo H, Volante D, Punzi L. Effects counts. Development and validation in a prospective longitudinal study of
of an educational-behavioural joint proctection program on people with patients with rheumatoid arthritis. Arthritis Rheum. 1995;38:44–8.
moderate to severe rheumatoid arthritis: a randomized controlled trail. Clin 33. Terslev L, Qvistgaard E, Kristoffersen H, Topr-Pedersen, Bliddal H. Doppler
Rheumatol. 2007;26:2043–50. ultrasonography in rheumatic diseases. Ugeskr Laeger. 2004;166:371–4.
14. Danish Society of Rheumatologists. Reumatoid artrit - klinisk retningslinje 34. Ellegaard K, Torp-Pedersen S, Terslev L, Danneskiold-Samsøe B, Henriksen M,
(Rheumatoid Arthritis-Clinical Guideline). https://danbio-online.dk/ Bliddal H. Ultrasound colour Doppler measurements in a single joint as
vejledning/dansk-reumatologisk-selskabs-kliniske-retningslinje/drs-kliniske- measure of disease activity in patients with rheumatoid arthritis -
retningslinje-for-diagnostik-klassifikation-behandling-og-monitorering-af-ra. assessment of concurrent validity. Rheumatology (Oxford). 2009;48:254–7.
15. Amris K, Wæhrens EE, Jespersen A, Bliddal H, Danneskiold-Samsøe B. 35. Naredo E. Ultrasonographic assessment of inflammatory activity in
Observation-based assessment of functional ability in patients with chronic rheumatoid arthritis: comparison of extended versus reduced joint
widespread pain: a cross-sectional study. Pain. 2011;152:2470–6. evaluation. Clin Exp Rheumatol. 2005;23:881–4.
16. Amris K, Waehrens EE, Stockmarr A, Bliddal H, Danneskiold-Samsoe B. 36. Brown AK, Quinn MA, Karim Z, Conaghan PG, Peterfy CG, Hensor E,
Factors influencing observed and self-reported functional ability in women Wakefield RJ, P J O’C, Emery P. Presence of significant synovitis in
with chronic widespread pain: a cross-sectional study. J Rehabil Med. 2014; rheumatoid arthritis patients with disease-modifying antirheumatic drug-
46:1014–21. induced clinical remission: evidence from an imaging study may explain
17. Amris K, Waehrens EE, Christensen R, Bliddal H, Danneskiold-Samsoe B. structural progression. Arthritis Rheum. 2006;54:3761–73.
Interdisciplinary rehabilitation of patients with chronic widespread pain: 37. Hammer HB, Bolton-King P, Bakkeheim V, Berg TH, Sundt E, Kongtorp AK,
primary endpoint of the randomized, nonblinded, parallel-group IMPROvE Haavardsholm. Examination of intra and interrater reliability with a new
trial. Pain. 2014;155:1356–64. ultrasonographic reference atlas for scoring of synovitis in patients with
18. Sreerangaiah D, Grayer M, Fisher BA, Ho M, Abraham S, Taylor PC. rheumatoid arthritis. Ann Rheum Dis. 2011;70:1995.
Quantitative power Doppler ultrasound measures of peripheral joint 38. Moher D, Hopewell S, Schulz KF, Montori V, Gotzsche PC, Devereaux PJ,
synovitis in poor prognosis early rheumatoid arthritis predict radiographic Elbourne D, Egger M, Altman DG. CONSORT 2010 explanation and
progression. Rheumatology (Oxford). 2016;55:89–93. elaboration: updated guidelines for reporting parallel group randomised
19. Taylor PC, Steuer A, Gruber J, Cosgrove DO, Blomley MJK, Marsters PA, trials. BMJ. 2010;340:c869.
Wagner CL, Maini RN. Comparison of ultrasonographic assessment of 39. Schulz KF, Altman DG, Moher D, CONSORT Group. CONSORT 2010
synovitis and joint vascularity with radiographic evaluation in a randomized, statement: updated guidelines for reporting parallel group randomised
placebo-controlled study of infliximab therapy in early rheumatoid arthritis. trials. BMJ. 2010;340:c332.
Arthritis Rheum. 2004;50:1107–16. 40. Fransen J, van Riel PL. The disease activity score and the EULAR response
criteria. Rheum Dis Clin N Am. 2009;35:745–57.
20. Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO 3 rd,
41. Ellegaard K, Torp-Pedersen S, Lund H, Pedersen K, Henriksen M,
Birnbaum NS, Burmester GR, Bykerk VP, Cohen MD, Combe B, Costenbader
Danneskiold-Samsoe B, Bliddal H. The effect of isometric exercise of the
KH, Dougados M, Emery P, Ferraccioli G, Hazes JM, Hobbs K, Huizinga TW,
hand on the synovial blood flow in patients with rheumatoid arthritis
Kavanaugh A, Kay J, Khanna D, Kvien TK, Laing T, Liao K, Mease P, Ménard
measured by color Doppler ultrasound. Rheumatol Int. 2013;33:65–70.
HA, Moreland LW, Nair R, Pincus T, Ringold S, Smolen JS, Stanislawska-
42. Holmgren T, Björnsson Hallgren H, Öberg B, Adolfsson L, Johansson K. Effect
Biernat E, Symmons D, Tak PP, Upchurch KS, Vencovský J, Wolfe F, Hawker
of specific exercise strategy on need for surgery in patients with
G;. Rheumatoid arthritis classification criteria: an American College of
subacromial impingement syndrome: randomised controlled study. Br J
Rheumatology/European League Against Rheumatism collaborative
Sports Med. 2014;48:1456–7.
initiative. Ann Rheum Dis 2010;69:1580–1588.
21. Ramey DR, Raynauld JP, Fries JF. The health assessment questionnaire 1992:
status and review. Arthritis Care Res. 1992;5(3):119–29. Review.
22. Rheumatoid arthritis in adults: management. Clinical guideline. nice.org.uk/
guidance/cg79. National Institute for Health and Care Excellence NICE . 15-
5-2017.
23. Heine PJ, Williams MA, Williamson E, Bridle C, Adams J, O'Brien A, Evans D,
Lamb SE. Development and delivery of an exercise intervention for
rheumatoid arthritis: strengthening and stretching for rheumatoid arthritis
of the hand SARAH trial. Physiotherapy. 2012;98:121–30.
24. Moritani T, deVries HA. Neural factors versus hypertrophy in the time course
of muscle strength gain. Am J Phys Med. 1979;58:115–30.
25. Gabriel DA, Kamen G, Frost G. Neural adaptations to resistive exercise: mechanisms
and recommendations for training practices. Sports Med. 2006;36:133–49.