2.6-Measurement Properties Flashcards
How is measurement used in Physiotherapy
- As a basis for prognosis, diagnosis and result evaluation
- Initial assessment
-Outcome assessment
Continuous Variable
-A variable along a scale within a defined range
-Examples: joint range, distance walked, time
Discrete Variable
-Whole unit variables
-Examples: Heart rate, number of steps
Dichotomous Variable
-Binary variable with a related context (qualitative)
-Example: Return to sport, survival
Indirect vs direct variables
Indirect- Correlation to a characteristic representative of the variable (Temperature, heart rate)
Direct- Observational variable (Height, weight)
Constructs (abstract variables)
Unobservable variables which are inferred based on relevant properties
(Abstract concepts-Motivation)
(Direct constructs- Temperature)
Limitation: Definition alters via discipline
Nominal
-Categorical with no particular order
- Mutually exclusive (cannot assign multiple categories, lowest yield)
- Blood type, handedness, pregnancy
Ordinal
-Ranked according to order in relation to a property
-Common in clinical assessment
-Non-quantitative relationship
-Sensation (Normal, impaired, absent)
Interval
-Rank order characteristics along with interval distances and units
-Relative difference can be determined
-Numerical value that does not represent true nature of the variable
-Temperature (0 zero temperature is not absence of heat or cold)
Ratio
-Empirical zero + interval scale
-Highest level of measurement
-Statistics can be applied
-Distance, age, time, weight, BP
Reliability
The degree to which individuals can be distinguished from
each other despite measurement error
Measurement error
he systematic and random error of an individual’s score that is
not attributed to true changes in the construct to be measured
Construct validity
The extent to which scores of the outcome measure relate to
other measures in a manner that is consistent with theoretically
derived hypothesis concerning the concepts that are being
measured
Criterion validity
The extent to which scores of the outcome measure relate to
the gold standard
Responsiveness
The ability of an outcome measure to detect clinically important
changes over time
Validity
A test must measure what it is ought to measure.
Systematic errors
-Predictable
-Constant and biased
-Occur in one direction at all times (under estimation or over estimation)
Random errors
-Non-predictable and due to uncontrollable factors such as fatigue, distraction and simple mistakes
-Error due to chance
Sources of error for patients?
- Cognitive impairment
- Physical impairment (pain, anxiety, fear etc)
- Lack of familiarity with procedure
- Distracted by environment
- Recovery or disease progression
How to minimise systematic error?
1.Test at similar times of the day
2. Check comprehension
3. Provide familiarisation of task to reduce nervousness and increase confidence
4. Standardise environment
5. Quiet environment
6. Assessment of pain intensity prior to test
7. Consider number of reps to avoid fatigue
Sources for error for clinicians
- Skill at performing the measurement, Preparation, Motivation / interest, Procedure used
2.Starting position
3.Handling - Instructions
- Data Collection (?standardized)
How to minimise systematic error for clinicians?
- Increase skill / competence of conducting / reading the measurement
- Procedure used
- Standardised starting position
- Standardised handling / stabilisation
- Standardised instructions
- Standardised use of equipment
- Standardised recording of results
Sources of error in equipment?
- Faulty Components
- Poor Calibration
- Variable Operating Conditions
- Poor Alignment / set up
- Requires Warm Up or Cool Down
How to minimise equipment error?
- Regular calibration / maintenance of equipment.
- Standardised warm-up time.
- Standardised operating conditions.
- Use the same piece of equipment.
Types of reliability?
- Test-retest
- Intra-rater
- Inter-rater
Test-retest reliability
To establish that an instrument is capable
of measuring a variable with consistency
Used where raters are not involved, such
as self-report questionnaires or in lab-
based assessments of equipment
Intra-rater reliability
- The stability of data recorded by one
individual across two or more trials - Same assessor, same subjects and across 2 trials.
Inter-rater reliability
- The consistency of measurements
between different clinicians when
measuring the same subject(s) - Multiple measurers, same subjects, same trial fashion
MCID- Minimal clinically important difference
is the smallest change that is
perceived to be beneficial by the
patient and would lead to a change in
management in the absence of
excessive side effects and costs
Distribution based approaches:
Degree of change in groups of patients
Related to a distribution of scores
(differences in group means and variance within the distribution)
E.g. standard response mean (SRM)
Anchor-based methods
Use external clinical criteria (anchor) to define clinical important
change
(e.g. a little better, a lot better, a little worse, a lot worse)