Authors:

Chieko Miyata, MD
Tetsuya Tsuji, MD, PhD Cancer
Akira Tanuma, MD
Aiko Ishikawa, MD
Kaoru Honaga, MD, PhD
Meigen Liu, MD, PhD
ORIGINAL RESEARCH ARTICLE
Affiliations:
From the Department of Rehabilitation
Medicine, Keio University School of
Medicine, Tokyo, Japan (CM, TT, AI,
KH, ML); and Division of Rehabilitation Cancer Functional Assessment Set
Medicine, Shizuoka Cancer Center,
Shizuoka Prefecture, Japan (CM, AT). A New Tool for Functional Evaluation in Cancer
Correspondence:
All correspondence and requests for
ABSTRACT
reprints should be addressed to: Miyata C, Tsuji T, Tanuma A, Ishikawa A, Honaga K, Liu M: Cancer functional
Tetsuya Tsuji, MD, PhD, Department of assessment set: a new tool for functional evaluation in cancer. Am J Phys Med
Rehabilitation Medicine, Keio
University School of Medicine, Rehabil 2014;93:656Y664.
Shinanomachi 35, Shinjuku-ku, Tokyo, Objective: The aim of this study was to develop and evaluate the psycho-
160-8582, Japan.
metric properties of a new measurement tool, the Cancer Functional Assess-
Disclosures: ment Set (cFAS).
Supported in part by the National Design: A total of 119 inpatients with cancer participated in this prospective
Cancer Center Research and
cohort study. This study took place in three phases: (1) item generation, literature
Development Fund (23-A-29).
Presented at the Multinational search, and a round-table discussion by 32 rehabilitation specialists; (2) item
Association for Supportive Care in reduction and selection of appropriate items from the first item pool; and (3)
Cancer (MASCC) symposium in Berlin
on June 29, 2013 (poster presentation). evaluation of psychometric properties. Concurrent validity (Spearman correlation
Financial disclosure statements have coefficients with existing scales), internal reliability (Cronbach coefficient alpha),
been obtained, and no conflicts of
interrater reliability (weighted kappa statistics and intraclass correlation coefficient),
interest have been reported by the
authors or by any individuals in control cross-validation, and responsiveness (standardized response mean values) were
of the content of this article. examined.
Results: The cFAS consisted of 24 items. Significant correlations were found
0894-9115/14/9308-0656 between the cFAS and existing scales. Cronbach > for the total score was 0.92.
American Journal of Physical
Weighted J values for each item ranged from 0.74 to 1.00. The intraclass cor-
Medicine & Rehabilitation
Copyright * 2014 by Lippincott relation for the total score was 0.97. Concurrent validity and internal consistency
Williams & Wilkins were similar at two different hospitals. The cFAS was more responsive to changes
than the other tools.
DOI: 10.1097/PHM.0000000000000082
Conclusions: The cFAS has acceptable psychometric properties, supporting
its broad generalizability. It can be a useful tool in clinical trials and can contribute to
the development of cancer rehabilitation.
Key Words: Rehabilitation, Malignancy, Physical Function, Instrument, Impairment

656 Am. J. Phys. Med. Rehabil. & Vol. 93, No. 8, August 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
D uring the past several years, quality-of-life
has become a serious concern in the treatment of
patients with cancer. Cancer screening for early de-
tection greatly increased the chance for successful
treatment, and the choice of treatment options avail-
able to cancer patients has expanded. Multimodal
treatments combining surgery with chemotherapy,
such as molecular-targeted drugs and radiotherapy,
have improved both overall and disease-free survival.
However, these treatments occasionally involve acute,
chronic, and delayed treatment toxicities. A regimen
that considers treatment benefits, complications, and
side effects needs to be developed. Moreover, quality-
of-life and activities of daily living are key aspects
that also need to be considered in decisions about
cancer care and treatment. Because most cancer pa-
tients have impairment and disability during various
stages of their illness, the demand for cancer rehabil-
itation, from the early period of diagnosis to the ter-
minal stage, is increasing. Previous investigations have
indicated that rehabilitation significantly improves the
motor functions and cognitive functions of cancer
patients. Induced by both exacerbation of the disease
and side effects of treatment, various changes occur in
physical functions and abilities. Rehabilitation spe-
cialists must evaluate the neuromuscular, musculo-
skeletal, and functional status of the patients properly
to devise more effective training programs for im-
proving their status. Such a measurement must be a
single task assessment that is both usable in daily
clinical practice and suitable for various stages of
cancer. Furthermore, the measurement must be able
to detect changes in a timely fashion after they occur.
To date, the Eastern Cooperative Oncology
GroupYPerformance Status (ECOG-PS),1 the Karnofsky
Performance Scale (KPS),2 the Palliative Perfor-
mance Scale (PPS),3 and the Edmonton Functional
Assessment ToolY2 (EFAT-2)4 have been developed
as physical function scales of cancer patients. Using
these scales, the general condition and performance
status of cancer patients can be understood. These
are also used to determine whether patients can re-
ceive chemotherapy and as a prognostic tool for
survival in palliative care. However, when used in
cancer rehabilitation, these have the following limi-
tations: (1) discrimination between impairment
and disability of the patients is not clear, (2) some
evaluation standards lack objectivity, (3) the evalua-
tion standard may not be able to assess the small
changes of physical function, and (4) these cannot adapt
to the recent changes in cancer treatment strategies
such as the need for hospitalization. Therefore, it is
necessary to develop a new tool to measure physical
www.ajpmr.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
functions that is easy to administer, short, reliable,
valid, and responsive to clinical changes during re-
habilitation interventions. The purpose of this study
was to develop and confirm the psychometric prop-
erties of the Cancer Functional Assessment Set (cFAS),
a new tool for evaluating physical functions in cancer
patients.
METHODS
Participants
The participants were 119 cancer inpatients
at varying stages of illness. Sixty-nine inpatients
were recruited from June 2009 to February 2010 at
Shizuoka Cancer Center (SCC), and 50 inpatients
(20 for interrater reliability testing and 30 for
cross-validation study) were recruited from August
2010 to August 2011 at Keio University Hospital
(KUH). The inclusion criteria were (1) patients
who received cancer treatment and (2) patients who
underwent physiotherapy and/or occupational ther-
apy. The exclusion criteria were (1) patients who
had difficulty in communicating and various other
conditions judged unsuitable for entry by the at-
tending physician and (2) patients who underwent
rehabilitation for functional deficits not related to
cancer and its treatment.
Study Design
Development and validation of the cFAS took
place in three phases: phase 1, item generation;
phase 2, item reduction; and phase 3, evaluation of
psychometric properties. The study was approved
by the Ethical Committee of SCC and KUH accord-
ing to the guidelines of the Helsinki declaration, and
informed consent was obtained from the patient.
Phase 1: Item Generation
Systematic reviews related to cancer rehabili-
tation were retrieved from MEDLINE and EMBASE
from December 2003 to December 2008. Guyatt
et al.5 have highlighted the importance of consen-
sus expert opinion as it relates to evidence-based
medicine. A round-table discussion was under-
taken, and 32 board-certified physiatrists with a
minimum of 10 yrs’ experience in rehabilitation
medicine were asked to list factors relevant to im-
pairments and disabilities of cancer patients. When
developing the cFAS, a minimum number of items
required for the evaluation of impairments in
daily clinical practice, a single task assessment, and
Using cFAS to Measure Functionality 657
 suitability for various stages of cancer were the basic
principles considered.
Phase 2: Item Reduction
The strategy for item reduction was guided by a
desire to ensure item relevance and fit with patient
values. Reactivity to functional changes and evalu-
ation practicality in cancer patients were particu-
larly important. The data of 45 inpatients at SCC in
whom repeated evaluation was possible were used.
Items that showed good response to changes or
were clinically important were retained. The stan-
dardized response mean (SRM),6 the ratio between
the mean change score and the standard deviation
of that change score within the same group, was
calculated to assess the responsiveness of each item.
An SRM of approximately 0.20 is considered small,
whereas 0.50 indicates moderate responsiveness,
and greater than 0.80 is considered highly sensi-
tive.6 Items with low responsiveness and those that
most patients could not perform were excluded.
Phase 3: Psychometric Properties of the cFAS
The data of 69 inpatients at SCC were used to
assess concurrent validity and internal reliability
of the cFAS. Concurrent validity of the cFAS was
assessed by correlating the total cFAS score with
the other performance tools (ECOG-PS, KPS, and
Functional Independence Measure [FIM] instru-
ment7), measured at the same time, using Spearman
correlation coefficients. Cronbach alpha was calcu-
lated to determine the degree of internal reliabil-
ity of the 24 items of the cFAS. This coefficient is
considered acceptable when it is greater than 0.7,
according to Nunnally8 criterion.
Interrater reliability testing was conducted in
20 inpatients at KUH. Two raters evaluated each
patient with the cFAS independently within a 24-hr
period. To estimate interrater reliability, weighted
kappa statistics were used for the 24 items, and
the intraclass correlation coefficient (ICC; 2,1) was
used for the total cFAS score. Weighted kappa sta-
tistics are frequently used for scale items that con-
tain more than two possible responses to more
accurately reflect the amount of disagreement
among observers. According to conventional cut-
offs for interpreting kappa statistics, reliability
was interpreted as almost perfect (0.81Y1.00), sub-
stantial (0.61Y0.80), or moderate (0.41Y0.60).9 An
ICC of 0.7 is commonly used as a threshold of ac-
ceptable reliability.10
Another 30 inpatients at KUH were recruited
for the cross-validation study. The same rater as-
sessed the cFAS at two different hospitals, SCC and
KUH, to examine concurrent validity and internal
consistency.
As additional evaluation, the responsiveness
of the total cFAS score was assessed. The SRM was
used to evaluate the responsiveness of the total
cFAS score, the ECOG-PS, the KPS, and the FIM
instrument. Physical function was measured either
before rehabilitation intervention or once weekly
after the initiation of rehabilitation. A data set of
75 inpatients in whom repeated evaluation was
possible at both hospitals was used.
The Statistical Package for the Social Sciences
for Macintosh, version 18.0 (Statistical Package for
the Social Sciences Inc, Chicago, IL), was used for
all statistical analyses, setting the significance level
at less than 5%.

TABLE 1 Patients’ characteristics

Concurrent Validity and

Internal Reliability Interrater Reliability Cross-validation

Total SCC KUH KUH

N 119 69 20 30
Sex
Female 45 (37.8%) 24 (34.8%) 9 (45%) 12 (40%)
Male 74 (62.3%) 45 (65.2%) 11 (55%) 18 (60%)
Age, yrs 64.1 (11Y85) 61.9 (11Y85) 67.7 (34Y81) 66.8 (33Y85)
Primary tumors
Gastrointestinal 33 21 5 7
Lung 20 13 3 4
Hematologic 15 8 1 6
Brain 15 13 0 2
Genitourinary 10 1 3 6
Head and neck 9 5 3 1
Other 17 8 5 4

658 Miyata et al. Am. J. Phys. Med. Rehabil. & Vol. 93, No. 8, August 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
RESULTS
Table 1 shows the characteristics of the 119
patients (37.8% women, 62.2% men; mean age,
64.1 yrs).
Phase 1: Item Generation
Eight instruments1Y4,7,11Y13 were identified in
the systematic review. Those instruments served as
a framework from which to guide an expert con-
sensus on cancer rehabilitation. The initial item
pool contained 41 items that were commonly af-
fected in cancer patients and considered funda-
mental to functional performance. These included
shoulder abduction passive range of motion (ROM),
knee extension passive ROM, ankle dorsiflexion pas-
sive ROM, grip strength, iliopsoas manual muscle
testing (MMT),14 quadriceps MMT, tibialis anterior
MMT, abdominal muscle strength of the Stroke
Impairment Assessment Set,15 the finger floor dis-
tance, upper extremity sensory function, lower ex-
tremity sensory function, sitting balance, body sway
with feet together and eyes closed for 1 min, one-foot
standing, Timed Up & Go test,16 the number of times
the patient stood up in 30 secs, upper extremity
function of the Stroke Impairment Assessment Set,
rolling over, sitting up, standing up, transfer, 50-m
walk, going up and down the stairs, locomotion, areas
of daily activities, communication (understanding
and expression), intellectual function (MiniYMental
State Examination17), motivation for rehabilitation,
and consciousness state (Glasgow Come Scale18).
Phase 2: Item Reduction
At first, each item chosen in phase 1 was graded
from 0 to 3 or 0 to 5. A rating of 0 represents a total
loss of function, whereas a rating of 3 or 5 repre-
sents optimal function. The following items whose
SRM level was less than 0.50 were excluded: knee
extension passive ROM, sitting balance, knee-
mouth test of the Stroke Impairment Assessment
Set, finger function test of the Stroke Impairment
Assessment Set, rolling over, locomotion, commu-
nication, motivation for rehabilitation, and con-
sciousness state. Although the responsiveness of the
lower and upper extremity sensory function items
was small (0.26 and 0.43), these were not excluded.
This is because these are valuable items for cancer
patients who have neural deficits as a result of
treatment complications and direct invasion of the
tumor. Items that were judged to be inappropriate
for cancer patients were the finger floor distance,
the Timed Up & Go test, the number of times the
patient stood up in 30 secs, and intellectual function.
www.ajpmr.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Their achievement quotients were less than 0.90.
The remaining 24 items constituted the cFAS
(Table 2).
Sit up/stand up/transfer/walk/stairs ascending
and descending are basic activities of daily living.
For the sit up, stand up, transfer, 50 m-walk, and
stairs ascending and descending items, the amount
of assistance scored is purely regarding the physi-
cal assistance given to the patient and does not
take into account other factors such as number of
steps climbed or distance walked. Grip strength
is also a marker of physical fitness and muscle
strength. The MMT of the iliopsoas and the abdo-
men can screen for steroid myopathy. The quadri
ceps muscle strength is important for basic activi-
ties. Sensory function and the MMT of the tibialis
anterior are useful for peroneal neuropathy. Body
sway with feet together, eyes closed for 1 min, is
an indicator of position sense. The one foot stand-
ing with eyes open task is a test of static balance.
Ankle dorsiflexion passive ROM with knee flexed
can assess the tightness of the triceps surae and
the heel cord. Shoulder abduction passive ROM is
able to pick up the functional disorder such as ad-
hesive capsulitis of the shoulder.
Phase 3: Evaluating Psychometric
Properties
The Spearman correlation coefficients between
the total cFAS score and the ECOG-PS, the KPS,
and the FIM instrument were j0.79 (P G 0.0001),
0.75 (P G 0.0001), and 0.73 (P G 0.0001), respectively.
Cronbach alpha was 0.92 for the total score, indi-
cating sufficient reliability according to Nunnally
criterion. Using weighted kappa for scale items,
agreement among investigators ranged from sub-
stantial (0.74) to almost perfect (1.00) (Table 3). The
ICC (2,1) for the total cFAS score was 0.97 and had
acceptable reliability.
There were no significant differences in the
mean age, sex, and primary tumor type between
samples of the two hospitals (Table 1). There were
also no significant between-sample differences in
the mean total cFAS score (SCC, 66.83 [17.39];
KUH, 67.27 [19.72]). Cronbach > was 0.91 for the
total scale, indicating sufficient reliability for re-
search purposes. The Spearman correlation co-
efficients between the total cFAS score and the
ECOG-PS, the KPS, and the FIM instrument were
j0.82 (P G 0.0001), r = 0.85 (P G 0.0001), and 0.86
(P G 0.0001), respectively.
Table 4 shows the SRM for the total cFAS score,
the ECOG-PS, the KPS, and the FIM instrument.
Using cFAS to Measure Functionality 659
 660
TABLE 2 Cancer Functional Assessment Set

Miyata et al.
Item 0 1 2 3 4 5
Sit up Maximal to Moderate assistance Minimal assistance Supervision Requires some device Independence
total assistance
Stand up Maximal to Moderate assistance Minimal assistance Supervision Requires some device Independence
total assistance
Transfer Maximal to Moderate assistance Minimal assistance Supervision Requires some device Independence
total assistance
50-m walk Maximal to Moderate assistance Minimal assistance Supervision Requires some device Independence
total assistance
Stairs ascending and Maximal to Moderate assistance Minimal assistance Supervision Requires some device Independence
descending of one floor total assistance
Grip strength (Rt Lt) in the sitting G10 kg 10Y15 kg 15Y20 kg 20Y25 kg 25Y30 kg 930 kg
position with the elbow extended
Iliopsoas MMT (Rt Lt) No palpable Palpable/observable Able to actively Able to actively Able to maintain Able to maintain
Quadriceps MMT (Rt Lt) or observable muscle contraction move through the move through the the test position the test position
Tibialis anterior MMT (Rt Lt) muscle contraction full available ROM full available ROM against gravity and against gravity and
with gravity eliminated against gravity moderate resistance maximal resistance
One foot standing Impossible 1Y2 secs 3Y4 secs 5Y6 secs 7Y9 secs Q10 secs
with eyes open (Rt Lt)
Body sway with feet Impossible Severe marked Moderate oscillations Slight oscillations
together, eyes closed for 1 min oscillation 910 cm 5Y10 cm G5 cm
Abdominal MMTa Unable to sit up Able to sit up Able to come to the Able to sit up against
provided that sitting position despite considerable
there is no resistance pressure on the sternum resistance
to the movement by the examiner
Shoulder abduction G140 degrees 140Y165 degrees 165Y175 degrees 9175 degrees
passive ROM (Rt Lt)
Ankle dorsiflexion passive G5 degrees 5Y15 degrees 15Y25 degrees 925 degrees
ROM with knee flexed (Rt Lt)
U/E sensory function Severe dysfunction Moderate dysfunction Minimal dysfunction Normal
(including light touch,
position, and abnormal sensation)

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
L/E sensory function (including light Severe dysfunction Inhibits function Minimal dysfunction Normal
touch, position, and abnormal sensation) moderately
The area where he/she is doing On the bed In the room In the house/in the unit Outdoors/in
daily activity the hospital
a
The patient rests in the 45-degree semireclining position in a wheelchair or high-back chair and is asked to raise the shoulders off the back of the chair and assume a sitting position.
L/E, lower extremity; Lt, left; Rt, right; U/E, upper extremity.

Am. J. Phys. Med. Rehabil. & Vol. 93, No. 8, August 2014
 TABLE 3 Weighted kappa and ICC of the cFAS
Weighted J
Abdominal MMT 0.98
Sit up 0.95
Stand up 0.95
Transfers: bed/chair/wheelchair 0.98
Gait 0.99
Stairs going up and down 0.96
Activity area 0.98
Rt grip strength 0.97
Lt grip strength 0.85
Rt iliopsoas MMT 0.74
Lt iliopsoas MMT 0.97
Rt quadriceps MMT 0.79
Lt quadriceps MMT 0.92
Rt tibialis anterior MMT 0.86
Lt tibialis anterior MMT 0.86
Standing with feet together, eyes closed and hands by the sides 0.87
Rt one foot standing 0.95
Lt one foot standing 0.89
Rt ankle dorsiflexion passive ROM 0.96
Lt ankle dorsiflexion passive ROM 0.95
Upper extremity sensory function 1.00
Lower extremity sensory function 0.95
Rt shoulder abduction passive ROM 0.96
Lt shoulder abduction passive ROM 0.97
Lt, left; Rt, right.

The SRM was higher for the cFAS than for the other
rating scales, regardless of the treatment regimen.
DISCUSSION
Need for an Instrument to Assess
Physical Functions of Cancer Patients
One of the goals of this study was to develop the
cFAS (Table 2), a new tool for measurement of
physical functions in phase 1 and 2 cancer patients.
Cancer rehabilitation involves helping cancer pa-
tients obtain optimal physical, social, psychologic,
and vocational functioning within the limits im-
posed by the disease and its treatment.19 Physical
functions must be evaluated in detail to offer ap-
propriate rehabilitation, and rehabilitation goals
need to be modified in consideration of physical
functioning and patients’ prognosis. Objective data
also play a pivotal role as a common language for
healthcare professionals, and these must be useful
in multidisciplinary care. Furthermore, cancer re-
habilitation can be better demonstrated to medical
professionals of other specialized fields if the effects
of rehabilitation can be shown by an objective
physical function scale.
As physical function scales of cancer patients,
the ECOG-PS, the KPS, the PPS, and the EFAT-2
have been developed to date. The ECOG-PS and the

TABLE 4 Responsiveness according to treatment (SRM)

Scale

Treatment Regimen cFAS ECOG-PS KPS FIM

Symptomatic therapy (including palliative care) (n = 24) 1.01 0.81 0.80 0.87
Radical surgery (n = 6) 1.26 0.61 0.61 0.89
Palliative surgery (n = 8) 0.98 0.33 0.33 0.76
Chemotherapy (n = 14) 0.94 0.68 0.73 0.61
Radiotherapy (n = 11) 0.95 0.32 0.44 0.80
Palliative surgery + chemotherapy (n = 4) 0.95 0.72 0.56 0.84
Concurrent chemoradiation therapy (n = 8) 1.01 0.57 0.57 0.86
Total (N = 75) 0.93 0.61 0.64 0.75

www.ajpmr.com Using cFAS to Measure Functionality 661

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 KPS have been used widely to assess both how a
patient’s disease is progressing and how the disease
affects the patient’s daily living abilities, as well as to
determine the appropriate treatment and prognosis.
The KPS, developed by Karnofsky et al.2 in 1948,
ranges from 100 to 0, where 100 is normal func-
tioning and 0 is death. In 1960, the ECOG intro-
duced a simpler scale, similar to the KPS, with only
5 points. This is now termed the ECOG-PS, having
been expanded to consist of 6 points with the ad-
dition of PS 5. More recently, the PPS and the
EFAT-2 have been developed as functional measures
for use in a palliative care population. The PPS has
been modified from the KPS to assess ambulation,
activity, self-care, intake, and consciousness level.
The EFAT includes domains such as pain, mental
alertness, sensory function, communication, and
respiratory function, in addition to physical function.
However, the previous scales fall short as a
comprehensive instrument to assess impairments
of cancer patients. These are not designed to eval-
uate cancer patients’ physical conditions based on
either the International Classification of Impair-
ments, Disabilities, and Handicaps20 or the Interna-
tional Classification of Functioning, Disability and
Health21 model. Recently, the International Classifi-
cation of Functioning, Disability and Health core sets
for cancer have been developed, but these are limited
to only breast cancer and head and neck cancer.13,22
Even though the ECOG-PS, the KPS, and the PPS are
simple and help one to easily grasp the patient’s
general condition, these cannot describe why the
patient’s activities are limited. It has also been
suggested that the scores on the ECOG-PS and the
KPS for patients with terminal cancer had a Bfloor
effect.[4 Although the PPS and the EFAT-2 are scales
that are specialized for end-of-life care, their evalua-
tion criteria are vague.
Because cancer rehabilitation involves adapta-
tion in various stages of the illness, a measure for
patients with cancer must be usable in various
stages of the illness. To get more insight into the
physical status and to plan more effective treatment
programs for patients with cancer in accordance to
their physical conditions, impairments and dis-
abilities must be assessed.
The newly developed cFAS assesses impair-
ments in daily clinical practice without the need for
a special tool. In the present study, the SRMs of the
ECOG-PS, the KPS, and the FIM instrument were
uneven, depending on the treatment regimens, but
the SRM of the cFAS maintained the same respon-
siveness across the regimens at a higher level than
the other scales. The cFAS can be considered a
662 Miyata et al. Am. J. Phys. Med. Rehabil. & Vol. 93, No. 8, August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
highly detailed scale that can evaluate the physical
function changes of patients with cancer, regardless
of treatment regimen, during various stages.
Psychometric Properties of the cFAS
The second goal of this study was to evaluate
the psychometric properties in phase 3. The mea-
surement tool had to possess the following char-
acteristics: reliability, validity, internal consistency,
responsiveness, and practicality.23 Reliability refers
to the reproducibility of a measurement. Validity
is an index as to how well any evaluation tool ac-
tually measures what it intends to measure. Internal
consistency is the extent to which a test assesses
the same characteristics. Responsiveness refers to
the ability of an evaluation tool to detect clinically
relevant changes.
The strong correlations of the cFAS with the
ECOG-PS, the KPS, and the FIM instrument dem-
onstrate its concurrent validity with the existing
scales that have been used to measure the effects of
cancer rehabilitation. The Cronbach alpha coeffi-
cient for the cFAS exceeded Nunnally criterion of
0.7, which demonstrates acceptable internal con-
sistency of the cFAS. Concerning the interrater re-
liability, weighted kappa statistics for the 24 items
of the cFAS and ICC for the total cFAS score dem-
onstrated sufficient reliability (Table 3). These re-
sults indicate that the estimates of the interrater
reliability of the cFAS are acceptable.
Furthermore, cross-validation was performed.
In an independent sample, it was confirmed that
Cronbach alpha coefficients for the total cFAS were
acceptable and that the cFAS had strong correla-
tions with the existing scales. The newly developed
cFAS can be performed in daily clinical practice; it
is based on single-task assessment, and it can be
completed in 5Y10 mins. Furthermore, it does not
require special training to perform the assessment
because the evaluation standard is clear.
These results indicate that the cFAS is a reli-
able, valid, responsive, and practical instrument.
Furthermore, the cross-validity of the cFAS was
confirmed in another sample.
Advantages of the cFAS
The psychometric properties discussed above
indicate that the cFAS can be very useful in daily
clinical practice, as well as in clinical research.
Its advantages over other existing instruments
are the following:
(1) It is developed on the basis of the World Health
Organization model of International Classification
of Impairments, Disabilities, and Handicaps and
 therefore shows physical function definitely. It is
useful to plan more effective rehabilitation pro-
grams and change the goal of rehabilitation inter-
vention in accordance with a patient’s condition.
(2) It is usable to evaluate any cancer patient.
(3) It is based on single-task assessment and can be
performed in daily clinical practice.
(4) It does not require special training to perform
the assessment because the evaluation standard
is clear.
(5) It is sufficiently responsive to changes.
(6) It is responsive regardless of the treatment
regimen and appropriate for clinical application
to various cancer patients.
This study, however, suggested room for further
improvement of the cFAS. The cFAS cannot evaluate
physical fitness because it was developed as a scale
evaluable in daily clinical practice. It also does not
describe the symptoms (physical and psychologic
complaints) of the patients. The authors will try to
evaluate these symptoms using an additional scale.
The cFAS with the symptom scale might be more
useful for outcome studies in cancer rehabilitation. It
also needs to be validated in not only the inpatient but
also the outpatient population. Lastly, the limited
sample size makes it difficult to draw broader con-
clusions on the basis of tumor types and resulting
impairments.
CONCLUSIONS
The cFAS meets most of the requirements for
use in oncology clinical trials, including ease of
administration, brevity, reliability, validity, and re-
sponsiveness to important clinical changes. Fur-
thermore, it is superior to existing scales as a tool
for, and a potential contributor to the development
of, cancer rehabilitation.
ACKNOWLEDGMENTS
The authors thank Professor Hideyuki Saya of
the Division of Gene Regulation, Institute for Advanced
Medical Research, Keio University School of Medicine,
for kind advice and encouragement throughout
this study. The authors also thank Professor Toru
Takebayashi of the Department of Preventive Medicine
and Public Health, Keio University School of Medicine,
and Takayuki Abe, assistant professor of Biostatistics,
Keio University School of Medicine, for their guidance
and advice in relation to statistical analyses.
REFERENCES
1. Zubrod CG, Schneiderman M, Frei E, et al: Appraisal
of methods for the study of chemotherapy of cancer
www.ajpmr.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
in man: Comparative therapeutic trial of nitrogen
mustard and triethylene thiophosphoramide. J Chron
Dis 1960;11:7Y33
2. Karnofsky DA, Abelmann WH, Craver LF, Burchenal
JH: The use of nitrogen mustards in the palliative
treatment of cancer. Cancer 1948;1:634Y56
3. Anderson F, Downing M, Hill J, et al: Palliative Per-
formance Scale (PPS): A new tool. J Palliat Care
1996;12:5Y11
4. Kaasa T, Wessel J: The Edmonton Functional As-
sessment Tool: Further development and validation
for use in palliative care. J Palliat Care 2001;17:
5Y11
5. Guyatt G, Vist G, Falck-Ytter Y, et al: An emerg-
ing consensus on grading recommendations? ACP J
Club 2006;144:A8Y9
6. Bradley J, Howard J, Wallace E, et al: Reliability,
repeatability, and sensitivity of the modified shuttle
test in adult cystic fibrosis. Chest 2000;117:
1666Y71
7. Deutsch A, Braun S, Granger C: The Functional
Independence Measure and the Functional Inde-
pendence Measure for Children: Ten years of de-
velopment. Crit Rev Phys Med Rehabil 1996;8:
267Y81
8. Nunnally JC: Psychometric Theory. New York,
McGraw-Hill, 1978
9. Lyden PD, Lau GT: A critical appraisal of stroke
evaluation and rating scales. Stroke 1991;22:1345Y52
10. Hripcsak G, Heitjan DF: Measuring agreement in
medical informatics reliability studies. J Biomed In-
form 2002;35:99Y110
11. Mahoney FI, Barthel DW: Functional evaluation; the
Barthel index. Md State Med J 1965;14:61Y5
12. Cella DF, Tulsky DS, Gray G, et al: The Functional
Assessment of Cancer Therapy scale: Development
and validation of the general measure. J Clin Oncol
1993;11:570Y9
13. Brach M, Cieza A, Stucki G, et al: ICF core sets for
breast cancer. J Rehabil Med 2004;44:121Y7
14. Daniels L, Worthingham C: Muscle Testing: Tech-
nique of Manual Examination, ed 5. Philadelphia,
PA, WB Saunders Co, 1986
15. Chino N, Sonoda S, Domen K, et al: Stroke Impair-
ment Assessment Set (SIAS): A new evaluation in-
strument for stroke patients. Jpn J Rehabil Med
1994;31:119Y25
16. Podsiadlo D, Richardson S: The timed Bup & go[: A
test of basic functional mobility for frail elderly per-
sons. J Am Geriatr Soc 1991;39:142Y8
17. Folstein MF, Folstein SE, McHugh PR: BMini-mental
state[. A practical method for grading the cognitive
state of patients for the clinician. J Psychiatr Res
1975;12:189Y98
18. Teasdale G, Jennett B: Assessment of coma and im-
paired consciousness. A practical scale. Lancet 1974;
2:81Y4
Using cFAS to Measure Functionality 663
 19. Cromes GF: Implementation of interdisciplinary can-
cer rehabilitation. Rehabil Couns Bull 1978;21:230Y7
20. World Health Organization: International Classifi-
cation of Impairments, Disabilities, and Handicaps
(ICIDH). Geneva, Switzerland, World Health Orga-
nization, 1980
21. World Health Organization: World Health Organi-
zation International Classification of Functioning,
Disability and Health. Geneva, Switzerland, World
Health Organization, 2001
22. Tschiesner U, Rogers S, Dietz A, et al: Development of
ICF core sets for head and neck cancer. Head Neck
2010;32:210Y20
23. Liu M, Chino N, Tsuji T, et al: Psychometric prop-
erties of the Stroke Impairment Assessment Set
(SIAS). Neurorehabil Neural Repair 2002;16:339Y51

664 Miyata et al. Am. J. Phys. Med. Rehabil. & Vol. 93, No. 8, August 2014

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.