See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/256660353

Introducing Clinimetrics

Article in The Australian journal of physiotherapy December 2005

DOI: 10.1016/S0004-9514(05)70019-2

CITATIONS READS

9 54

1 author:

Mary Galea
University of Melbourne
228 PUBLICATIONS 5,359 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ESWT as a treatment for Neurogenic Heterotopic Ossification View project

Assessing body composition at bedside in acute spinal cord injury using bioelectrical impedance
analysis View project

All content following this page was uploaded by Mary Galea on 22 August 2015.

The user has requested enhancement of the downloaded file.


Introducing Clinimetrics
Mary P Galea
Rehabilitation Sciences Research Centre, School of Physiotherapy, The University of Melbourne
In this issue of the Journal we introduce a new feature
Clinimetrics.
A characteristic of good clinical practice is that it uses
measurement instruments that are reliable, valid, and
responsive to intervention. The role of measurement in
physiotherapy practice cannot be overstated. Indeed the APA
Position Statement on Clinical Justification and Outcome
Measures (2003) has as its first sentence: The progressive
evaluation of physiotherapy treatment outcomes is an integral
part of professional accountability and is a requirement of the
Australian Physiotherapy Competency Standards
19942003. Our profession has, on the whole, moved away
from a tendency to use assessment procedures simply because
they are available; now assessment procedures are more often
chosen because they are reliable and valid. Using valid and
reliable measures greatly increases the likelihood of being
able to measure accurately the impact of an intervention or
change over time.
Physiotherapy intervention is directed towards changing
symptoms that have complex constructs (for example, pain or
spasticity), and these constructs must be defined
operationally before they can be measured. Pain, for example,
has sensory, cognitive and emotional dimensions (Melzack et
al 1982) that are not represented in many of the measurement
tools used to assess pain in routine clinical practice. Spasticity
can be defined operationally as a velocity-dependent increase
in the stretch reflex (Lance 1980) or as resistance to passive
movement (Bohannon and Smith 1987). The operational
definition will determine how this construct is measured: the
Tardieu scale (Fosang et al 2003) is based on Lances
definition, whereas the Modified Ashworth Scale is based on
the latter definition.
It is also important to understand the assumptions of
measurement theory and its limitations. Many instruments
used in physiotherapy practice involve the use of a scale for
scoring purposes. Scaling is the assignment of objects to
numbers according to a rule and development of a scale
involves assumptions about whether the underlying construct
is unidimensional or multidimensional. While many scales
are similar in that they each measure the concept of interest
on a line, they may differ considerably in how scale values are
derived for different items. For example, both the Likert and
Guttman scales involve the respondent indicating agreement
with certain statements, but the method of computing the
score is quite different in each case. Practically all
measurement of human behaviours involves errors.
Understanding the nature and source of these errors can help
in reducing their impact and may in some instances prevent
the drawing of incorrect conclusions. Understanding how a
measurement tool was developed, including the study
population and the size of the sample used, is important as
Australian Journal of Physiotherapy 2005 Vol. 51
Editorial
this will influence its generalisability.
As we become more sophisticated in understanding
mechanisms underlying movement dysfunction it may be
possible to classify subgroups of patients, and therefore
diagnostic scales will be used (Petersen et al 2004). The
sensitivity and specificity of these scales need to be
understood in order draw meaningful conclusions about their
utility as highlighted in recent Critically Appraised Papers in
this Journal (Cumming 2000, Riddle 2001)
The purpose of Clinimetrics is to alert clinicians to the
psychometric properties of instruments that have clinical
utility in current physiotherapy practice and in recent
research. To achieve this purpose the Editorial Board has
adopted procedures to identify instruments used to measure
prognosis, diagnosis, outcome, and economics of health
problems managed by physiotherapists. These instruments
are then summarised in a structured format that describes the
instrument, the method of scoring or deriving the variable
used, the target population(s), its validity, its reliability (or
major sources of error or potential bias), its
sensitivity/specificity (for prognosis or diagnosis), and
clinically relevant change (for measuring outcome). A
commentary on the selected instrument is provided by
individuals with both clinical and research expertise in the
clinical problem addressed by the instrument and a sound
grasp of measurement theory. The commentary will include
an interpretation of how the test should be used. Each issue of
the journal will henceforth contain Clinimetrics
contributions. Because it is impossible to evaluate all
acceptable instruments, the Clinimetrics Editors will
endeavour to provide a mix of instruments relevant to a range
of sub-disciplines of physiotherapy.
Of course the idea of examining measurement instruments in
this way is not new. The Canadian Physiotherapy Association
first published a guide to physical rehabilitation outcome
measures in 1994 and recently published an updated version
(Finch et al 2002). The Victorian Neurology Special Interest
Group has also developed a manual of outcome measures
with annotations about the strengths and limitations of each
instrument (Hill et al 2002). These documents are of
enormous value to both clinicians and researchers, and it is
not the intention of the Editorial Board to duplicate them.
Rather, the intention is to present examples of measurement
tools used for different purposes across the spectrum of
physiotherapy practice and to highlight aspects of the
instruments that have considerable bearing on their clinical
utility.
Clinimetrics has been modelled on a similar feature in the
Dutch Physiotherapy journal Nederlands Tijdschrift voor
Fysiotherapie. Albere Koke developed the concept of
139
Editorial

Clinimetrics, which appears in the Dutch journal as Meten in Fosang A, Galea MP, McCoy AT, Reddihough DS and Story I
de Practijk. He and Els van den Ende, Editor of Nederlands (2003): Measures of muscle and joint performance in the lower
Tijdschrift voor Fysiotherapie, were supportive of our desire limb for children with cerebral palsy. Developmental Medicine
and Child Neurology 45: 664670.
to replicate this concept in the Australian Journal of
Physiotherapy. Jennifer Keating and Megan Davidson were Hill K, Miller K, Denisenko S, Clements T and Batchelor F (2002):
also involved in discussions about the format of this feature. Manual for Clinical Outcome Measurement in Adult
Neurological Rehabilitation (2nd ed). Melbourne: APA Victorian
We extend our sincere thanks to all these people for their Neurology Special Group Publication.
contribution to the final product.
Lance J (1980): Session synopsis. In: Feldman RG, Young RR
References and Koella WP (Eds) Spasticity: Disordered Motor Control.
Chicago: Year Book Medical Publishers pp 1724.
Australian Physiotherapy Association (2003): Position Statement
Melzack R, Wall PD and Ty TC (1982): Acute pain in an
on Clinical Justification and Outcome Measures. Melbourne:
emergency clinic: Latency of onset and descriptor patterns
Australian Physiotherapy Association.
related to different injuries. Pain 14: 3342.
Bohannon RW and Smith M (1987): Interrater reliability of a
Petersen T, Olsen S, Laslett M, Thorsen H, Manniche C, Ekdahl
Modified Ashworth Scale of muscle spasticity. Physical Therapy
C and Jacobsen S (2004): Inter-tester reliability of a new
67: 206207.
diagnostic classification system for patients with non-specific
Cumming B (2000): Positive forward bending tests for scoliosis low back pain. Australian Journal of Physiotherapy 50: 8591.
are usually false positives. Negative tests are usually true
Riddle D (2001) A set of three clinical tests can detect the
negatives. [Commentary] Australian Journal of Physiotherapy
presence or absence of rotator cuff tears. [Commentary]
46: 314.
Australian Journal of Physiotherapy 47: 298.
Finch E, Brooks D, Stratford P and Mayo N (2002): Physical
Rehabilitation Outcome Measures: A Guide to Enhanced
Clinical Decision Making (2nd ed). Toronto: Canadian
Physiotherapy Association.

140 Australian Journal of Physiotherapy 2005 Vol. 51

View publication stats