Understanding the role of stimulation in reflexology: development and testing of a robotic device.

Abstract

Reflexology is a common choice of women with breast cancer as supportive care during treatment. It involves stimulation of specific locations of the feet called reflexes using a specialised walking motion with the thumb of the reflexologist. Reflexology has shown potential for the successful management of cancer and treatment-related symptoms and improvement in physical functioning; however to date, the mechanism of action for these improvements is unknown. One confounder to the study of reflexology is the 'human factor'. To study the effects of the stimulation of the reflexes independent of the 'human factor', there is a need for an alternative method for the delivery of reflexology. The objective of this work was to design and create a robotic reflexology device that would deliver a breast cancer-specific reflexology protocol to the feet of patients. A prototype robotic reflexology device was developed and tested for feasibility, safety and acceptability with breast cancer survivors (n= 13), and preliminary efficacy in symptom management and enhanced functional status with a sample of women undergoing chemotherapy for breast cancer (n= 13). Safety, feasibility and acceptability were established, and significant improvements from pre- to post-device-delivered reflexology were seen in symptom severity among women on chemotherapy.

Full text

Understanding the role of stimulation in reflexology:
development and testing of a robotic deviceecc_1268 1..11
L.L. FLYNN, ms,graduate assistant,Department of Mechanical Engineering, College of Engineering, Michigan
State University, East Lansing, MI,T.R.BUSH,phd,assistant professor,Department of Mechanical Engineering,
College of Engineering, Michigan State University, East Lansing, MI, A. SIKORSKII, phd,assistant professor,
Department of Statistics and Probability, College of Natural Science, Michigan State University, East Lansing, MI,
R. MUKHERJEE, phd,professor,Department of Mechanical Engineering, College of Engineering, Michigan State
University, East Lansing, MI, & G. WYATT, rn,phd,professor,College of Nursing, Michigan State University,
East Lansing, MI, USA
FLYNN L.L., BUSH T.R., SIKORSKII A., MUKHERJEE R. & WYATT G. (2011) European Journal of Cancer
Care
Understanding the role of stimulation in reflexology: development and testing of a robotic device
Reflexology is a common choice of women with breast cancer as supportive care during treatment. It involves
stimulation of specific locations of the feet called reflexes using a specialised walking motion with the thumb
of the reflexologist. Reflexology has shown potential for the successful management of cancer and treatment-
related symptoms and improvement in physical functioning; however to date, the mechanism of action for
these improvements is unknown. One confounder to the study of reflexology is the ‘human factor’. To study
the effects of the stimulation of the reflexes independent of the ‘human factor’, there is a need for an alternative
method for the delivery of reflexology. The objective of this work was to design and create a robotic reflexology
device that would deliver a breast cancer-specific reflexology protocol to the feet of patients. A prototype
robotic reflexology device was developed and tested for feasibility, safety and acceptability with breast cancer
survivors (n=13), and preliminary efficacy in symptom management and enhanced functional status with a
sample of women undergoing chemotherapy for breast cancer (n=13). Safety, feasibility and acceptability were
established, and significant improvements from pre- to post-device-delivered reflexology were seen in
symptom severity among women on chemotherapy.
Keywords: reflexology,breast cancer,symptom management,robotic device.
INTRODUCTION
Breast cancer is the most commonly diagnosed cancer
among women, with more than 40 000 deaths annually
(American Cancer Society 2010). Women with breast
cancer are burdened with an array of symptoms from the
time of diagnosis, and throughout the treatment process.
Chemotherapeutic agents are linked to multiple symp-
toms including fatigue, weakness and insomnia. If these
symptoms are not properly managed, their presence can
lead to treatment interruptions and deterioration in func-
tioning, and may affect survival (Miaskowski et al. 2006;
Chang et al. 1998, 2007; Diedrich et al. 2007). Most
women consider various avenues of support to enhance
their health-related quality of life (HRQOL) as they con-
tinue through treatment, often turning to complementary
and alternative medicine (CAM) therapies (Kosachik et al.
2006). Boon et al. (2007) reported that 80% of women
Correspondence address: Tamara Reid Bush, Department of Mechanical
Engineering, College of Engineering, Michigan State University, East
Lansing, MI 48824, USA (e-mail: reidtama@egr.msu.edu).
Funding: The authors gratefully acknowledge the financial support pro-
vided by Michigan State University’s College of Engineering, College of
Nursing, the Division of Engineering Research and the Vice President for
Research and Graduate Studies.
Accepted 24 April 2011
DOI: 10.1111/j.1365-2354.2011.01268.x
European Journal of Cancer Care, 2011
Original article
© 2011 Blackwell Publishing Ltd

diagnosed with breast cancer turn to CAM for symptom
management. Among CAM therapies, reflexology is one
of the most commonly used (Eisenberg et al. 1998).
Lengacher et al. (2002) ranked reflexology in the top 10 in
frequency of use of traditional and ethnic medicines
among women with breast cancer.
Reflexology is a specialised foot therapy that is applied
primarily to the sole of the foot by a certified reflexologist.
The reflexologist uses a firm thumb-walking pressure over
specific regions referred to as reflexes. Reflexology is
similar to massage in that it manipulates soft tissue for
therapeutic purposes but also differs from massage due to
the specific focus on the feet and the use of a firm thumb-
walking motion on the reflexes (Snyder & Cheng 1998).
While the use of reflexology is extremely high (Eisen-
berg et al. 1993, 1998; Lengacher et al. 2002) and literature
on reflexology research is continuously expanding, mixed
evidence of its effectiveness is reported (Hodgson 2000;
Stephenson et al. 2003; Wang et al. 2008; Ernst 2009). One
systematic review by Ernst (2009) found that in three
cancer studies, reflexology resulted in a significant
decrease in anxiety among breast and lung cancer patients,
a second showed significant improvement in quality of
life among patients in palliative care and a third reported
no change in depression or anxiety in a sample of pallia-
tive care patients. A second systematic review by Wang
et al. (2008) that did not include cancer studies found a
decrease in urinary symptoms among multiple sclerosis
patients. A third review (Wilkinson et al. 2008) docu-
mented an immediate post-intervention reduction in pain.
Apart from the systematic reviews, five randomised
control trials (RCTs) focused on a cancer patient popula-
tion. Among the five studies, sample sizes ranged from 18
to 60 patients per group and provided one to eight sessions
of reflexology for the intervention group, with a range of
15–30 min per session. One study reported significantly
lower anxiety among multiple types of cancer patients
(Quattrin et al. 2006); another found lower pain among
patients with mixed cancers (Stephenson et al. 2003); a
third reported decreased pain among digestive cancer
patients (Pan et al. 2000); a fourth, using a bundled inter-
vention, found eight weekly sessions of reflexology plus
self-initiated support to significantly promote relaxation
among post-surgical early-stage breast cancer patients (n=
60) (Sharp et al. 2010); and a fifth study involved a mixed
sample of hospitalised cancer patients (n=42) and taught
a partner to conduct one session, resulting in an immedi-
ate decrease in pain intensity and anxiety (Stephenson
et al. 2007).
While multiple studies have investigated the effect of
reflexology on HRQOL of cancer patients, there is no
single scientifically established mechanism of action for
how reflexology affects HRQOL outcomes including
symptoms and function. One theory suggests that the
mechanism of action is that the nerve endings in the feet
connect with different areas of the body, and the direct
pressure on a specific nerve ending of the foot stimulates
the corresponding area of the body for symptomatic relief
(Hodgson 2000). According to this theory, specific areas of
the foot (reflexes) correspond to specific organs or systems
of the body (Fig. 1), and thus manipulation of these spe-
cific areas can lead to the improvement in symptoms.
A protocol that targets reflexes tied to symptoms of breast
cancer and its treatment has been developed (Wyatt et al.
2005).
In addition to the stimulation of specific reflexes, the
‘human factor’ is also thought to influence outcomes of
reflexology. The human factor includes the experience of
the reflexologist in finding and stimulating specific
reflexes. The interpersonal style of the reflexologist during
the session may also be a confounding human factor. In
addition, the human touch by itself may have a therapeu-
tic effect, and the touch produced by different reflexolo-
gists may be different (e.g. different motion patterns or
methods of force application). This variability in the deliv-
ery of reflexology that is due to human factor may explain
some of the variable effects on patient outcomes. Thus, to
understand the role of stimulation of the reflexes in
reflexology, removing the human factor is necessary. One
of the ways to remove the human factor is to stimulate the
reflexes using a mechanical device.
Currently available commercial foot devices that claim
to be ‘reflexology-based devices’ apply a generic vibration
over the entire foot, or random stimulation of non-specific
reflexes. None of the current devices provide the equiva-
lent of a hands-on reflexology session. Such a device needs
to apply and release pressure along specific paths on the
sole of the foot, for specific durations, in a consistent
fashion with control for force levels. In addition, the
device would need to follow a specific protocol associated
with select reflexes tailored for breast cancer. Commercial
devices do not exist with these attributes; therefore, there
was a need to develop a device with these specifications.
The purpose of this work was to begin to uncover the
specific components of reflexology which may or may not
play a role in symptom management and improvements in
physical functioning among cancer patients. Testing
against hands-on reflexology is planned in the future, as
the device may provide an alternative for some cancer
patients who are homebound. The intent of this line of
work is not to replace a therapist, but rather to advance
the science by isolating the key active ingredients of
FLYNN et al.
© 2011 Blackwell Publishing Ltd2

reflexology, and in the long run possibly offer in-home or
in-clinic options for some patients. However, large-scale
clinical testing in a randomised trial would be required
before the device-delivered reflexology could be offered to
patients as supportive care during cancer treatment.
Thus, the goals of this study were to:
Goal 1: Design and create a robotic reflexology device
that would provide a breast cancer-specific reflexology
protocol;
Goal 2: Test the device for feasibility and acceptability
with a sample of breast cancer survivors; and
Goal 3: Conduct a preliminary evaluation of the efficacy
of the device for symptom management and improvement
in physical function of women with breast cancer under-
going chemotherapy.
MATERIALS AND METHODS
Reflexology protocol
According to a breast cancer-specific reflexology protocol
tested in the ongoing RCTs of hands-on reflexology (Wyatt
et al. 2005), the reflexes associated with symptoms of
breast cancer and its treatment are located in three
regions: (1) the sole of the foot; (2) around the base of the
ankle on the top of the foot; and (3) at the base of the toes
on the top of the foot. During stimulation, the mechanical
device mimics the thumb-crawling motion of a reflexolo-
gist. The total time for the protocol is 15 min per foot.
Each region and stimulation of corresponding reflexes by
the device is discussed in detail below.
Stimulation of the sole of the foot
Seven reflexes are stimulated as discussed below.
Spine reflex The device begins by applying pressure
along the inside arch of the foot, spanning from the base of
the big toe to the heel. The robotic device is programmed
to identify the starting point and uses a walking motion
with pulsing pressure while moving from the heel to the
base of the great toe. According to one of the current
theories of reflexology, this reflex stimulates the nervous
system which is responsible for maintaining homeostasis
throughout the body. Homeostasis can be significantly
impaired by breast cancer and the associated therapies,
and this reflex assists with rebalancing (Watson & Voner
2009) (Fig. 1).
Lung and diaphragm reflexes The robotic reflexology
device then locates the region that runs from the base of
the toes to approximately the top one-third of the sole of
the foot (Figs 1 and 2). The device uses the walking pulsing
pressure and moves from the inside (great toe area) to the
outside (little toe) and reverses direction. Next, the device
automatically locates the diaphragm guideline and stimu-
lates along the line which is the diaphragm reflex. The
Figure 1. Reflexology chart for the sole
of foot – (chart location: http://www.
exhalereflexology.com/wp-content/
uploads/2008/09/reflexology-chart.jpg).
A robotic device for reflexology
© 2011 Blackwell Publishing Ltd 3

lung and diaphragm reflexes are included in the protocol
since they are believed to support the essential exchange
of oxygen and carbon dioxide with the environment
(Watson & Voner 2009).
Kidneys and adrenal reflexes The device locates the area
that spans the mid third region on the sole of the foot.
These reflexes are part of the protocol since the kidneys
remove wastes from the body and regulate water, salt and
nutrients in the blood. Since the adrenal glands are located
on top of the kidneys, this wider combined reflex stimu-
lates both organs. The adrenals are included in the proto-
col since they function to regulate the hormones in the
body that are often affected by cancer treatment (Watson
& Voner 2009). In addition to this broad stimulation, a
focused stimulation is conducted on an area approxi-
mately the size of a kidney bean that is located below the
diaphragm line, just lateral to the tendon line. The kidney
reflex is specifically stimulated to further support the
cleansing functions mentioned above.
Spleen reflex The robotic reflexology device focuses on
the stimulation of the region approximately the size of a
dime on the outer edge of the foot in the midsection of the
sole (Fig. 1). The reflex is stimulated since it contributes
to the production of red blood cell and antibodies and may
aid in the removal of cells damaged by cancer treatment
(Watson & Voner 2009).
Intestinal reflex The sole of the foot is stimulated
between the heel line and mid foot. The device moves
back and forth several times across this entire area using
the walking pulsing motion that mimics the motion of a
reflexologist. This reflex addresses the many intestinal
symptoms develop due to cancer treatment including both
diarrhoea and constipation.
Stimulation on top of foot
Breast and chest reflexes A mechanism, separate from
the one used for stimulation of the sole of the foot, is used
for stimulation of the region at the base of the toes which
encompasses the breast and chest reflexes. The mecha-
nism is positioned in the area that runs across the base of
the toes between the second and fifth toe (Fig. 3). It applies
a rolling pressure to the soft tissue between all toes simul-
taneously. The breast/chest reflex stimulation assists in
the lymphatic drainage which is often impaired after
breast cancer surgery (Byers 1996).
Stimulation of the base of the ankle
Lymphatic reflexes A crawling pulsing pressure horizon-
tally is produced across the ankle region starting from
under the lateral ankle bone and continuing across the
front of the foot to the medial ankle bone. This movement
Figure 2. (A) Robotic reflexology device,
exterior view. (B) Mechanism for sole
stimulation, showing the inside of the
robotic reflexology device. (C) Subject
using the device.
FLYNN et al.
© 2011 Blackwell Publishing Ltd4

stimulates the flow of lymph so that it is less likely to
become congested in the arm and surrounding tissue after
breast surgery (Byers 1996).
Design and development of the robotic device
The device has been designed to follow the breast cancer-
specific protocol described above.
Automated foot calibration
Because patients have a wide variety of foot sizes and
shapes, a semi-automated procedure was developed for
locating the reflexes with the device. By taking a few
landmark measurements of the foot, the device is able to
automatically identify the locations of the reflexes used in
the protocol. This method requires an initial session to
non-invasively take measurements of each foot, and enter
these measurements into the computer program operating
the device. These measurements only need to be entered
once and saved in the computer so they can be used for
subsequent sessions with the same patient.
Foot calibration begins with a planar view digital
picture of the patient’s foot. The foot is aligned at approxi-
mately the same angle as it would be in the machine, and
a scale is placed next to the foot for reference. This digital
image is then used to determine landmarks on the bottom
of the foot through the use of Didge Image Digitizing
Software (2008; Didge, Omaha, NE, USA). This software
allows definition of the base of the heel as a reference
point. From this reference point, the distances to other
foot landmarks are determined. For stimulation of the sole
of the foot, six landmarks are necessary. Using the refer-
ence image, the positions of the landmarks are selected on
the image and entered into the software. From these coor-
dinates the location of each reflex can be identified.
Since the device is programmable, different landmarks
or different paths for stimulation could be used to define
different reflexology protocols, if desired.
The device also has the ability to collect and store infor-
mation that includes time of usage, initial force/pressure
settings and any user changes made to the force/pressure
settings during the application of the reflexology protocol.
Mechanism for stimulation of the sole of the foot
The robotic reflexology device was designed as a rectan-
gular unit with a calf support (Fig. 2). The rectangular unit
has a central location for placement of the left or right
foot. The central location is an opening that is covered by
a thin, opaque, smooth and stretched, flexible fabric. The
robotic component/mechanism for stimulation of the sole
of the foot (Fig. 2B) is located inside the rectangular device
and below the central opening. This component is not
visible to the patient. The contact point, which applies the
pressures, is comprised of a small sphere mounted on a
motorised slide. The sphere applies pressure on the sole of
the foot and its movement is controlled by two other
slides so the sphere can travel in any direction. The flex-
ible fabric prevents direct contact between the foot and
the sphere, keeps friction constant between the two and
prevents any pinching between the foot and device.
Once the patient is positioned and her foot dimensions
are recorded by the device, the custom computer program
is started. The computer is programmed to automatically
determine the regions/locations on the sole of the foot to
be stimulated and time trajectories following the breast
cancer-specific reflexology protocol.
Most importantly, the system has been designed with
safeguards so that the patient is not harmed. The force
sensor that measures the level of pressure applied is con-
tinuously monitored to ensure that desired force levels are
never exceeded.
Mechanism for stimulation of the top of the foot
A separate mechanism is used for stimulation of the
region at the base of the toes on the top of the foot (i.e. the
Figure 3. Unit used for stimulation of the
top of the foot at the base of the toes.
A robotic device for reflexology
© 2011 Blackwell Publishing Ltd 5

area that runs across the base of the toes between the
second and fifth toe). This mechanism is placed on the
patient’s foot by a test assistant or the patient can be
shown how to do this herself. This device applies a rolling
pressure to the soft tissue between the toes simulta-
neously for a fixed duration.
The robotic component/mechanism for stimulation on
the top of the foot is comprised of a small CAM-driven
adjustable arm with multiple wheels (Fig. 3). The arm is
rotated and the stimulation wheels are lifted to allow the
patient to place her foot under this mechanism. Once the
foot has been positioned, the arms are lowered such that
the wheels are located on the soft tissue regions between
the second and fifth toe. The CAM mechanism causes the
wheels to move back and forth to stimulate the reflex.
Medial/lateral adjustments in the arms of the device
enable it to adapt to different foot sizes. Also, the arms can
be repositioned to accommodate either the left or the right
foot. Tension adjustments are located on either side of the
device and allow for various levels of pressure exerted by
the wheels onto the foot. Future plans are to integrate this
stimulation unit with the main device so all components
operate as one unit.
Stimulation of the base of the ankle
A third mechanism, mounted on the device (Fig. 2), is
wrapped around the base of the ankle. The ankle stimu-
lation mechanism consists of a series of small spheres
attached to a flexible cord. The woman wraps this
system around the base of her ankle. The clasp is a
spring/hook system which keeps tension on the spheres
while stimulation occurs. Once positioned, the woman
initiates stimulation through a remote control device.
The remote control turns on a motor that produces a
reciprocating motion. This in turn causes the spheres to
roll back and forth over the ankle region, stimulating the
reflexes. The woman also has the ability to select the
frequency of the reciprocating motion based on her level
of comfort.
Laboratory testing and oncology clinic testing
This research was approved by the university Human Sub-
jects Institutional Review Board and the participating
oncology clinic. Consent was obtained from all partici-
pants. This research involves two distinct samples:
•Sample 1: breast cancer survivors (n=13);
•Sample 2: women undergoing chemotherapy for breast
cancer (n=13).
Sample 1: laboratory testing
To test the device for safety, feasibility and acceptability
(goal 2), 13 breast cancer survivors were recruited.
The feasibility and acceptability testing were con-
ducted at the College of Engineering in the Biomechani-
cal Design Research Laboratory at Michigan State
University.
Sample 2: clinic testing
The preliminary efficacy testing (goal 3) was conducted
with a sample of women undergoing chemotherapy for
breast cancer (n=13). The device was located at a
community-based oncology clinic where women were
receiving chemotherapy for breast cancer.
Neither the laboratory nor the clinic setting provided
any special amenities (e.g., special lighting, music or
aromas), and women did not receive financial compen-
sation for their participation in the study. Women were
told the following during the consent process: ‘The
purpose of the study is to test a foot device that applies
therapeutic pressure to specific areas of the foot.
This research will test if the device helps you feel
better physically and/or emotionally during cancer
therapy’.
Intervention procedures
Women in both samples received four device-delivered
reflexology sessions – one per week for 4 weeks. All par-
ticipants were provided a pair of footie socks so that the
interaction with the device would be consistent across all
women and all sessions.
Data collection procedures
In sample 1 (breast cancer survivors), the first session was
preceded by paper and pencil collection of data including
demographics, symptoms and physical function data. This
was done while the foot calibration was performed by the
technician. After the fourth session with the device, data
were again collected via paper and pencil including symp-
toms, physical function, satisfaction and acceptability.
For sample 2 (women undergoing chemotherapy),
symptom and physical function data were collected via
telephone interviews at baseline and at 5 weeks (following
four weekly device sessions). If women adjusted the pres-
sure of the device during the sessions, these pressures
were recorded by the onboard computer.
FLYNN et al.
© 2011 Blackwell Publishing Ltd6

Measures
Demographics
Demographic data included age, race and ethnicity, level
of education and employment status.
Acceptability
Acceptability was assessed by a 13-question instrument
developed for this project. Women were asked to rate the
acceptability of session duration, device size, pleasantness
of and relaxation during session, absence of pain, degree of
stimulation, comfort of leg and foot positions, the appro-
priateness of chair height and the use of socks. Each item
was rated on the scale from 1 =completely unacceptable
to5=completely acceptable.
Symptom inventory (Cleeland et al. 2000)
The M. D. Anderson Symptom Inventory (MDASI) evalu-
ates severity of 13 symptoms experienced by cancer
patients (i.e. pain, fatigue, nausea, disturbed sleep,
distress, shortness of breath, difficulty remembering,
decreased appetite, drowsiness, dry mouth, sadness,
numbness/tingling), and the interference of these symp-
toms with daily life. Both severity and interference are
rated on a scale from 0 =absent to 10 =maximum pos-
sible. Reliability and validity of the instrument were
established (Cleeland 2007). Summed symptom severity
and interference scores were computed before and after
device sessions. The potential range of scores was 0–130
for severity, and 0–60 for interference with higher scores
indicating worse severity and interference.
Medical Outcomes Study Short Form 36 (MOS SF-36)
(Ware et al. 1996)
The physical function subscale of SF-36 has 10 items that
are scored using Likert’s method of summated ratings
(Ware et al. 1993). The instrument has established content
and construct validity and internal consistency reliability
(Ware et al. 1993, 1996). The potential range of scores was
0–100 with higher scores reflecting better function.
Force data
The force applied to the sole of the foot was measured
using a single-axis force transducer mounted directly
under the force applicator. This transducer allowed accu-
rate force measurement when the applicator was pressed
into the sole of the foot. The patient-controlled force dial
contained numbers from 0 to 10 indicating the level of
force. The controller was provided to the patient, so she
could increase or decrease the level of force depending on
her level of comfort and desired level of stimulation.
Data analysis
For both samples, descriptive statistics for demographic
characteristics of the participants, their symptom severity
and interference, physical function and responses to
acceptability questions were obtained. Mean values of the
force at each specific reflex for each foot were determined
for the purposes of setting them as defaults in the future,
and plotted over time (sessions 1–4) to determine if
women’s preferred force settings change as time
progresses. The trends in force levels over time were
analysed for the locations corresponding to the protocol
steps on right and left feet separately. Matched paired
t-tests were used to evaluate the change in symptoms and
function from before to after the device sessions. Since the
purpose of the study was to evaluate feasibility and
acceptability and obtain preliminary efficacy data, the
study was not powered for the formal tests of hypotheses.
All analyses were performed in sas 9.1 (SAS Software
2010; SAS Institute Inc., Cary, NC, USA).
RESULTS
Goal 1 and sample demographics
A robotic device for the delivery of a breast cancer-specific
reflexology protocol was successfully designed and built.
It was tested with both sample 1 and sample 2. In both
samples, over one-third of women were employed (part-
time or full-time), all had at least some college education,
and the majority were married or living with a partner
(Table 1). In sample 1, all 10 women completed all four
device sessions as prescribed by our study protocol.
Goal 2
In sample 1, the summary of acceptability ratings in
Table 2 revealed very high mean levels for: session dura-
tion (4.77), pleasantness (4.69), relaxation (4.77), absence
of pain during session (4.46), comfort of leg and foot (4.69),
comfort of chair height (4.85) and acceptability of socks
worn (5.00). Women were also asked to rate the accept-
ability of three device sizes for home use. Medium and
small sizes received the highest ratings and the current
size was rated at the mean of 3.08 (neutral).
Furthermore, in sample 1, women were breast cancer
survivors who were no longer on treatment and therefore
A robotic device for reflexology
© 2011 Blackwell Publishing Ltd 7

had few symptoms as reflected by the mean symptom
severity score of 0.96 (Table 3). The symptom severity and
interference and physical function did not change
(improved insignificantly) from pre- to post-device ses-
sions, further supporting feasibility and acceptability of
the robotic reflexology device.
Goal 3
In sample 2, the symptom severity significantly decreased
(P=0.02) from pre- to post-device sessions, and physical
function improved (P=0.06) from pre- to post-device ses-
sions. The effect sizes for these improvements expressed
as Cohen’s d(difference between means in standard devia-
tion units) were 0.86 for symptom severity and 0.70 for
physical function. In Cohen’s classification (Cohen 1988),
the cut-off for the large effect size is 0.80, and our results
point to the potential efficacy of the device with respect to
symptom management and improvement of physical
function during chemotherapy.
Furthermore, in sample 2, three women dropped out
after the intake interview, and did not complete the device
sessions and post-sessions interview. The other 10 women
out of 13 in sample 2 (77%) completed both interviews. Of
these 10 women, nine completed all four device sessions,
and one woman completed three out four. Thus, feasibil-
ity of device sessions held in the clinic was supported.
Additional results: level of force selected by women
Across all reflexes for sample 1, the levels of forces
decreased over time with varying magnitude, indicating
that women on average dialled back the levels at later
sessions compared to the earlier ones. For one of the loca-
tions (left ball of foot), the negative slope was significantly
different from zero; however, the sample size in this pilot
study was small to draw conclusions based on P-values
alone (Fig. 4).
DISCUSSION
Goal 1
A robotic device that applies a breast cancer-specific
reflexology protocol was successfully developed and
tested. This device is principally different from existing
‘foot massagers’ in several ways. The device can identify
reflex locations on a patient’s foot and can target specific
Table 1. Demographic characteristics of the participants in the study samples
Characteristic Category
Sample 1,
n(%)
Sample 2,
n(%)
Age, mean (SD) 61 (9.5) 56 (11.3)
Education Completed high school 0 (0) 1 (7.7)
At least some college 6 (46.2) 6 (46.2)
Completed college 3 (23.1) 5 (38.5)
Professional or graduate degree 4 (30.8) 1 (7.7)
Marital status Never married 3 (23.1) 1 (7.7)
Married/living with partner 8 (61.5) 8 (61.5)
Divorced 2 (15.4) 3 (23.1)
Widowed 0 (0) 1 (7.7)
Employment Full-time 4 (30.8) 4 (30.8)
Part-time 1 (7.7) 1 (7.7)
Not employed 0 (0) 3 (23.1)
Retired 6 (46.2) 2 (15.4)
Disabled 1 (7.7) 1 (7.7)
Homemaker 1 (7.7) 2 (15.4)
Race Caucasian/White 13 (100) 11 (84.6)
Black/African American 0 (0) 2 (15.4)
Table 2. Participants’ acceptability of the device and device ses-
sions (sample 1)
Mean (SD) on the scale
from 1 =completely
unacceptable to 5 =
completely acceptable
Acceptability of session duration 4.77 (0.44)
Potential device size: large 3.08 (1.44)
Potential device size: medium 3.23 (1.42)
Potential device size: small 4.61 (1.12)
Pleasantness of the session experience 4.69 (0.75)
Session experience was relaxing 4.77 (0.60)
Absence of pain during session 4.46 (1.71)
Session was not too stimulating 4.46 (1.85)
Leg and foot in a comfortable position 4.69 (0.63)
Comfort for left leg position 4.54 (1.20)
Comfort for right leg position 4.54 (1.20)
Appropriate chair height 4.85 (0.38)
Acceptability of socks 5.00 (0.00)
FLYNN et al.
© 2011 Blackwell Publishing Ltd8

reflexes rather than providing a generic stimulation (e.g. a
generic vibration or movement). The device is also pro-
grammable for time spent in stimulating a specific reflex
and the reflexes to be stimulated. Thus, even though the
current protocol is specific to symptoms of breast cancer
and its treatment, the protocol can be tailored to address
the unique concerns of patients with other cancers or
other diseases. The device separates the human factor
from the stimulation of specific reflexes. Allowing this
separation is the beginning to the uncovering of the
mechanism of action of reflexology on HRQOL including
symptom and function outcomes.
Goals 2 and 3
Data from preliminary testing indicate that the applica-
tion of reflexology through a robotic device was not only
acceptable to laboratory patients, but showed beneficial
trends in decreasing symptom severity and increasing
physical function among clinic patients receiving chemo-
therapy. The reduction in symptom severity from pre- to
post-device sessions was statistically significant (P=0.02,
Table 3). Thus, it is possible that the stimulation of the
reflexes alone could be beneficial as supportive care for
cancer patients undergoing treatment; however, the effi-
cacy conclusions can not be drawn from the present study.
The results obtained in our pilot sample of women under-
going treatment could be used to appropriately power a
larger study, which would control for extraneous influ-
ences via randomisation, and compare device-delivered
reflexology to hands-on reflexology and standard care for
symptom management.
Additional results
Level of force selected by women
These force data will be used as input for future device
modifications: default values will be set for each region
and the device will begin at this level for each session.
Thus, in future uses of the device this modification may
allow women to have fewer adjustments during sessions.
Limitations
As mentioned earlier, the evidence reported is prelimi-
nary. First, the study did not have a sufficiently large
sample size to produce reliable estimates of the effect of
the robotic device on symptom and function outcomes.
Second, this study did not have a control group. However,
this limitation is somewhat offset by the fact that studies
with cancer patients undergoing chemotherapy have
reported very small improvements over time for patients
Table 3. Symptom severity, symptom interference and physical function before and after the device sessions
Outcome
Sample 1 Sample 2
Before device
sessions
(n=12*)
After device
sessions
(n=13)
Difference
(before–after)
(n=12)
Before device
sessions
(n=13)
After device
sessions
(n=10)
Difference
(before–after)
(n=10)
Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
Mean (SD)
P-value
Symptom severity† 0.96 (1.55) 0.89 (1.24) 0.09 (1.00) 46.46 (31.22) 31.10 (23.83) 10.00 (11.59)
P=0.02
Symptom interference† 0.58 (1.31) 0.33 (0.66) 0.11 (0.27) 15.77 (13.62) 13.80 (17.47) 0.8 (11.83)
P=0.84
Physical function‡ 83.35 (26.4) 84.17 (27.1) -0.83 (6.34) 45.77 (23.97) 61.5 (29.35) -14.00 (20.11)
P=0.06
*One participant in sample 1 did not fill out symptoms and function part of the pre-session assessment.
†Higher values indicate higher severity or interference.
‡Higher values indicate better physical function.
Figure 4. Average force values in newtons for the entire subject
pool. The data are plotted over the 4-week session.
A robotic device for reflexology
© 2011 Blackwell Publishing Ltd 9

in the control group. Given et al. (2004) observed the
effect sizes below 0.20 for the improvements in symptom
severity under usual care. In the ongoing RCT of hands-on
reflexology, women show no improvements in symptom
severity and physical function in the usual care control
group (Wyatt et al. 2005). Third, it is also possible that the
introduction of a new device may create a sense of expect-
ancy for improvement by the patient. Future RCT work
would need to compare this device against a device that
does not provide the stimulation of reflexes. Clearly, there
remains additional testing to be carried out with various
comparison groups and with larger samples.
The evidence of feasibility, safety and preliminary effi-
cacy presented here indicates that this approach to sup-
portive care for women with breast cancer during
chemotherapy warrants further examination.
Potential benefits
Applying reflexology through an objective means such as
a robotic device allows for repeatable and consistent
administration of the protocols. By using such a device,
the variability encountered across reflexologists and
within a reflexologist during a hands-on reflexology
session is removed.
In addition to the contribution to the science of reflex-
ology, there are other potential gains from the develop-
ment of a robotic reflexology device. Hands-on reflexology
requires a patient to travel outside the home for weekly
reflexology sessions. This places scheduling burdens on
patients who already are undergoing breast cancer treat-
ment in addition to meeting their daily needs. A robotic
device, such as the one developed for this research, has the
potential to offer supportive care to patients through an
accessible tool for managing symptoms during chemo-
therapy. Two usage possibilities were identified. Clinic
usage was explored in the current study with sample 2.
The advantage of having the device in the clinic would
come from the reduction in the number of trips for the
patient out of the home, since the sessions can be sched-
uled in accordance with patient’s visits to the oncology
setting. The other usage possibility is at home including
two possible scenarios. In the first scenario, the patient
could purchase a device and keep it permanently in her
home for usage at any time. In a second scenario, the
patient could rent a device for use in her home for a period
of time when symptom management is most needed.
These possibilities could be further investigated to expand
the pool of evidence-based supportive care interventions
during cancer treatment.
REFERENCES
American Cancer Society (2010) Breast
Cancer Facts and Figures. Available
at: http://www.cancer.org/acs/groups/
content/@nho/documents/document/f86
1009final90809pdf.pdf (last accessed 28
June 2011).
Boon H.S., Olatunde F. & Zick S.M.
(2007) Trends in complementary/alter-
native medicine use by breast cancer
survivors: comparing survey data from
1998 and 2005. BMC Women’s Health
7, 4. doi: 10.1186/1472-6874-7-4. Avail-
able at: http://www.biomedcentral.com/
content/pdf/1472-6874-7-4.pdf (accessed
23 October 2007).
Byers D.C. (1996) Better Health with Foot
Reflexology: The Original Ingham
Method. Ingham Publishing, St Peters-
burg, FL, USA.
Chang V.T., Thaler H., Polyak T., Kornblith
A., Lepore J.M. & Portenoy R.K. (1998)
Quality of life and survival: the role of
multidimensional symptom assessment.
Cancer 83, 173–179.
Chang Y., Lee J., Lee W., Lee K., Bang S.,
Wang X., Mendoza R., Cleeland C. & Yun
Y. (2007) Assessment of clinical relevant
fatigue level in cancer. Supportive Care
in Cancer 15, 891–896.
Cleeland C.S. (2007) Symptom burden:
multiple symptoms and their impact as
patient-reported outcomes. Journal of the
National Cancer Institute. Monographs
37, 16–21.
Cleeland C.S., Mendoza T., Wang X.S.,
Chou C., Harle M., Morrissey M. & Eng-
strom M. (2000) Assessing symptom dis-
tress in cancer patients: the M.D.
Anderson symptom inventory. Cancer
89, 1634–1646.
Cohen J. (1988) Statistical Power Analysis
for the Behavioral Sciences, 2nd edn.
Erlbaum, Hillsdale, NJ, USA.
Didge Image Digitizing Software (2008)
Version 2.3, 2008 Alistair Cullum,
Creighton University, Omaha, NE, USA.
Available at: http://biology.creighton.
edu/faculty/cullum/Didge/index.html
(last accessed 28 June 2011).
Diedrich J., Depke J. & Engel J. (2007) What
every nurse needs to know about breast
cancer: an overview and update of breast
cancer screening, diagnosis, and treat-
ment. American Nurse Today 2, 32–38.
Eisenberg D., Kessler R., Foster C., Norlock
F., Calkins D. & Delbacno T. (1993)
Unconventional medicine in the United
States: prevalence, costs, and patterns of
use. The New England Journal of Medi-
cine 328, 246–252.
Eisenberg D., Davis R., Ettner S., Appel S.,
Rompany M. & Kessler R. (1998) Trends
in alternative medicine uses in the
United States, 1990–1997. The Journal of
the American Medical Association 280,
1569–1575.
Ernst E. (2009) Is reflexology an effective
intervention: a systematic review of
randomized controlled trials. Medical
Journal of Australia 191, 263–266.
Given C., Given B., Rahbar M., Jeon
S., McCorkle R., Cimprich B., Galecki
S., Kozachik S., Brady A., Fisher-
Malloy M.J., Courtney K. & Bowie
E. (2004) Effect of a cognitive behav-
ioral intervention on reducing symp-
tom severity during chemotherapy.
Journal of Clinical Oncology 22, 507–
516.
Hodgson H. (2000) Does reflexology impact
on cancer patients’ quality of life?
Nursing Standard 14, 33–38.
Kosachik S., Wyatt G.K., Given C.W. &
Given B. (2006) Patterns of use of comple-
mentary therapies among cancer patients
and their family caregivers. Cancer
Nursing 29, 84–94.
Lengacher C., Bennett M., Kip K., Keller R.,
LaVance M., Smith L. & Cox C. (2002)
Frequency of use of complementary and
alternative medicine in women with
FLYNN et al.
© 2011 Blackwell Publishing Ltd10

breast cancer. Oncology Nursing Forum
29, 1445–1452.
Miaskowski C., Cooper B.A., Paul S.M.,
Dodd M., Lee K., Aouizerat B., West C.,
Cho M. & Bank A. (2006) Subgroups of
patients with cancer with different
symptom experiences and quality-of-life
outcomes: a cluster analysis. Oncology
Nursing Forum 33, E79–E89.
Pan C.X., Morrison R.S., Ness J., Fugh-
Berman A. & Leipzig R. (2000) Comple-
mentary and alternative medicine in the
management of pain, dyspnea, and
nausea and vomiting near the end of
life: a systematic review. Journal of
Pain and Symptom Management 20,
374–387.
Quattrin R., Zanini A., Buchini S., Turello
D., Annunziata M., Vidotti C., Colom-
batti A. & Brusaferro S. (2006) Use of
reflexology foot massage to reduce
anxiety in hospitalized cancer patients in
chemotherapy treatment: methodology
and outcomes. Journal of Nursing Man-
agement 14, 96–105.
SAS Software (2010) Version 9 of the
SAS system for Windows. Cary, NC:
Copyright 2002–2003 SAS Institute Inc.
SAS and all other SAS Institute Inc.
product or service names are registered
trademarks or trademarks of SAS Insti-
tute Inc.
Sharp D.M., Walker M.B., Chaturvedi A.,
Upadhyay S., Hamid A., Walker A.,
Bateman J., Braid F., Ellwood K., Hebble-
white C., Hope T., Lines M. & Walker L.
(2010) A randomized, controlled trial of
the psychological effects of reflexology in
early breast cancer. European Journal of
Cancer Care 46, 312–322.
Snyder M. & Cheng W.Y. (1998) Massage.
In: Complementary/Alternative Thera-
pies in Nursing, 3rd edn (eds Snyder M. &
Lindquist R.), pp. 285–293. Springer Pub-
lishing Company, New York, NY, USA.
Stephenson N., Dalton J.A. & Carlson J.
(2003) The effect of foot reflexology on
pain in patients with metastatic cancer.
Applied Nursing Research 16, 284–286.
Stephenson N., Swanson M., Dalton J.,
Keefe F.J. & Engelke M. (2007) Partner-
delivered reflexology: effects on cancer
pain and anxiety. Oncology Nursing
Forum 34, 127–132.
Wang M.Y., Tsai P.S., Lee P.H., Chang W.Y.
& Yang C.M. (2008) The efficacy of
reflexology: systematic review. Journal of
Advanced Nursing 62, 512–520.
Ware J.E., Snow K.K., Kosinski M. &
Gandek B. (1993) SF-36 Health Survey:
Manual and Interpretation Guide. The
Health Institute, New England Medical
Center, Boston, MA, USA.
Ware J.E., Kosinski M. & Keller S.S. (1996)
A 12-item short-form health survey: con-
struction of scales and preliminary tests
of reliability and validity. Medical Care
34, 220–233.
Watson S.A. & Voner V. (2009) Practical
Reflexology: Interpretation and Tech-
niques. McGraw-Hill, New York, NY,
USA.
Wilkinson S., Lockhart K., Gambles M. &
Storey L. (2008) Reflexology for symptom
relief in patients with cancer. Cancer
Nursing 31, 354–360.
Wyatt G.K., Sikorskii A. & Rahbar H.
(2005–2010) Reflexology: an intervention
for advanced breast cancer: Michigan
State University funded by the National
Cancer Institute. 1R01 CA104883-01A1.
A robotic device for reflexology
© 2011 Blackwell Publishing Ltd 11