Measuring pain
Summary
Karoly (1985) - we should focus on all of the factors that contribute to pain
1. Sensory - intensity, duration, threshold, tolerance, location, etc
2. Neurophysiological - brainwave activity, heart rate, etc
3. Emotional and motivational - anxiety, anger, depression, resentment, etc
4. Behavioural - avoidance of exercise, pain complaints, etc
5. Impact on lifestyle - marital distress, changes in sexual behaviour
6. Information processing - problem solving skills, coping styles, health beliefs
Techniques used to collect data.
1. interviews - advantage - it can cover Karoly's 6 points
2. behavioural observations
3. psychometric measures
4. medical records
5.
physiological measures
Physiological measures of pain
Muscle tension is associated with painful conditions such as headaches
and lower backache, and it can
be measured using an electromyograph (EMG). This apparatus
measures electrical activity in the muscles, which is a sign of how tense they
are. Some link has been established between headaches and EMG patterns, but EMG
recordings do not generally correlate with pain perception (Chapman et al 1985) and EMG measurements
have not been shown to be a useful way of measuring pain.
Another approach has been to relate pain to autonomic arousal. By taking measures of pulse rate, skin
conductance and skin temperature, it may be possible
to measure the physiological arousal caused by experiencing pain. Finally, since pain is
perceived within the brain, it may he possible to measure
brain activity, using an electroencephalograph
(EEG), in order to determine the extent to which an individual is
experiencing pain. It has been shown that subjective reports of pain do
correlate with electrical changes that show up as peaks in EEG recordings.
Moreover, when analgesics are given, both pain report and waveform amplitude on
the EEG are decreased (Chapman et al, 1985).
Evaluation
The advantage of the physiological measures of pain described above is
that they are objective (that is, not subject to bias by the person whose pain
is being measured, or by the person measuring the pain). On the other hand,
they involve the use of expensive machinery and trained personnel. Their main
disadvantage, however, is that they are not valid (that is, they do not measure
what they say they are measuring). For example, autonomic arousal can occur in
the absence of pain — being
wired up to a machine may be stressful and can cause a person’s heart rate to
increase. If someone is very anxious about the process of having his or her
pain assessed, or else is worried about the meaning of the pain, this will
cause physiological changes not necessarily related to the intensity of the
pain being experienced. Autonomic responses can be affected by many other
factors such as diet, alcohol consumption and infection. E.g. infection present
can get increased pulse rate. Better used as a signal for the presence of pain
rather than as a direct indices of pain.
Observations of pain
behaviours
People tend to behave in
certain ways when they are in pain; observing such behaviour could provide a
means of assessing pain.
Turk, Wack and Kerns (1985) have provided a classification of
observable pain behaviours.
• Facial /audible expression of distress: grimacing and teeth
clenching; moaning and sighing.
• Distorted ambulation or posture: limping or walking with a
stoop; moving slowly or carefully to protect an injury; supporting, rubbing or
holding a painful spot; frequently shifting position.
• Negative affect: feeling irritable; asking
for help in walking, or to be excused from activities; asking questions like
‘Why did this happen to me?’
• Avoidance of activity: lying down
frequently; avoiding physical activity; using a prosthetic device.
One way to assess pain behaviours is to observe them in a clinical
setting (although pain is also assessed in a natural setting as the patient
goes about his or her everyday activities). Keefe and Williams (1992) have
identified five elements that need to be considered when preparing to assess
any form of behaviour through this type of observation.
• A rationale for observation: it is
important for clinicians to know why they are observing pain behaviours. One
reason is to identify ‘problem’ behaviours that the patient may be reluctant to
report, such as pain when swallowing, so that treatment can be given. Another
is to monitor the progress of a course of treatment.
• A method for sampling pain behaviour techniques for sampling and
recording behaviour include continuous observation, measuring duration (how
long the patient takes to complete a task), frequency counts (the number of
times a target behaviour occurs) and time sampling (for example, observing the
patient for five minutes every hour).
• Definitions of the behaviour: observers need to be
completely clear as to what behaviours they are looking for.
• Observer training: in most clinical situations,
there will be different observers at different times and it is
important that they are consistent.
• Reliability and validity: the most useful measure of
consistency in observation methods is inter-rater reliability, but test-retest
reliability can also be useful. Three types of validity that could be assessed
are: concurrent validity (are the results of the observation consistent with
another measure of the same behaviour?), construct validity (are the
behaviours being recorded really signs of pain?) and discriminant validity (do
the observation records discriminate between patients with and without pain?).
A commonly used example of
an observation tool for, assessing pain behaviour is the UAB Pain Behaviour Scale designed by Richards et al (1982). This scale consists of ten target behaviours and observers
have to rate how frequently each occurs. 
The UAB is easy to use and
quick to score; it has scored well on inter-rater and test-retest
reliability.
However, correlation between scores on the UAB
and on the McGill Pain Questionnaire is low indicating that the relationship
between observable pain behaviour and the self-reports of the subjective
experience of pain is not a close one. Turk et al (1983) describe techniques
that someone living with the patient (the observer) can use to provide a
record of their pain behaviour. These include asking the observer to keep a
pain diary, which includes a record of when the patient is in pain and for how
long, how the observer recognized the pain, what the observer thought and felt
at the time, and how the observer attempted to help the patient alleviate the
pain. Other techniques are to interview the observer, or to ask the observer to
complete a questionnaire containing questions about how much the pain interferes
with the patient’s normal activities and social life, the effect of the pain on
family relationships and on the moods of both patient and observer.
Commentary
• Behavioural assessment is less objective
than taking physiological measurements, because it relies on the observer’s interpretation
of the patient’s pain behaviours (although, in practice, this can be partly
dealt with by using clearly defined checklists of behaviour and carrying out
inter-rater reliability — that
is, using two independent observers
and comparing their findings).
• An individual may be displaying a great deal
of pain behaviour, not because that individual is in severe pain but because
he or she is receiving social reinforcement for the pain behaviour (for
example, attention, sympathy and time off work). A by Gil et al (1988) provides an example of this: the children whose pain
behaviour (scratching their eczema) was rewarded with attention exhibited more
of this behaviour.
Self-report measures
Because pain is a subjective, internal experience, the assessment of
pain is therefore best carried out by using patient self-reports, and this is
by far the most frequently used technique.
Carroll (1993a) lists the different dimensions
of pain that sufferers can be questioned about:
• Site of pain: where
is the pain?
• Type of
pain: what does the pain feel like?
• Frequency of
pain: how often does the pain occur?
• Aggravating
or relieving factors: what makes the pain better or worse?
• Disability: how does the pain affect the patient’s everyday life?
• Duration
of pain: how long has the pain been present?
• Response to
current and previous treatments: how effective have
drugs and other treatments been?
An important item to add to this list is the emotional and cognitive effect
of the pain—in other words, how does the pain make patients feel and how does
it affect their thought processes and attitudes?
Two types of
self-rating scales
1. Visual
analogue
Patients mark a continuum of
severity from "No Pain" to "Very Severe Pain"
Simple and Quick to
use and can be filled out repeatedly
Can track the pain
experience as it changes - this could reveal patterns such as situations or
times of the day when the pain is better or worse
This method has
adequate reliability, however limits pain to a single dimension. Downie and colleagues evaluated
the degree of agreement between various scales in patients with rheumatic
diseases and found a high correlation among the different types of scales. The
scales are simple to understand and do not demand a high degree of literacy or
sophistication on the part of the patient, unlike other pain measurement tools,
such as the semantic differential scales described below. The Visual Analogue
Scale is simple and quick to administer, and may be used before, during, and
following treatment to evaluate changes in the patient's perception of pain
relative to treatment. The scales may also be completed throughout the course
of a day to assess change in pain intensity relative to activity or time of day.
2. McGill Pain Questionnaire (MPQ)
The McGill Pain Questionnaire, developed by Melzack
(1975), was the first proper self-report pain-measuring instrument and is still
the most widely used today.
An attempt to find words to describe
experiences of pain was made in a study by Melzack and Torgerson (1971) in
which they asked doctors and university graduates to classify 102 adjectives
into groups describing different aspects of pain. As a result of this exercise,
they identified three major psychological dimensions of pain:
• sensory: what the pain feels like physically —where it is located, how
intense it is, its duration and its
quality (for example, ‘burning’, ‘throbbing’)
• affective: what the pain feels like emotionally —whether it is frightening, worrying
and so on
• evaluative: what the subjective overall intensity of the pain experience is (for
example, ‘unbearable’, ‘distressing’).
Each of the three main classes was divided into a number of sub-classes (sixteen
in total). For example, the affective class was sub-divided into tension
(including the adjectives ‘tiring’, ‘exhausting’), autonomic (including
‘sickening’, ‘suffocating’) and fear (including ‘fearful’, ‘frightful’,
‘terrifying’).
Melzack and Torgerson (1971) then asked a sample
of doctors, patients and students to rate the words in each sub-class for
intensity. The first 20 questions on the McGill Pain Questionnaire consist of
adjectives set out within their sub-classes, in order of intensity. Questions 1
to 10 are sensory, 11 to 15 affective, 16 is evaluative and 17 to 20 are miscellaneous.
Patients are asked to tick the word
in each subclass that best describes their pain. Based on this, a pain rating index (PRJ) is calculated:
each sub-class is effectively a verbal rating scale and is scored accordingly
(that is, 1 for the adjective describing
least intensity, 2 for the next one and so on). Scores are given for the
different classes (sensory, affective, evaluative and miscellaneous), and also
a total score for all the sub-classes. In addition, patients are asked to
indicate the location of the pain on a body chart (using the codes E for pain
on the surface of the body, I for internal pain and El for both external and
internal), and to indicate present pain intensity (PPJ) on a 6-point verbal
rating scale. Finally, patients complete a set of three verbal rating scales
describing the pattern of the pain.
Criticism of this questionnaire centres on the need
to have extensive understanding of the English language eg discriminate between
words such as "Smarting" and "Stinging"
Semantic differential scales, such as the McGill, are difficult and time
consuming to complete and demand a sophisticated literacy level, a sufficient
attention span, and a normal cognitive state. They therefore are less
convenient to use in the clinical environment, but have value when a more
detailed analysis of a patient's perception of pain is needed, as in a pain
clinic or clinical research setting.
The issue of reliability has been addressed in numerous reports, particularly as it concerns the VAS and the McGill Pain Questionnaire. These reports do not lead to a consensus on reliability of these measurements. They suggest that reliability varies based on the patient groups that were examined for pain. Reliability therefore becomes an issue of "reliable in whose hands?" Reliability of many of the pain measurement methods have not extended in any realistic way beyond the reliability found by the original authors of the pain measurement methods.
A lack of clear reliability information should not prevent the clinician from using these methods, but it should alert the clinician to the possibility that a particular method may not be reliable with a particular patient or a group of patients. The clinician also should ensure that those who use the measurements for their own purposes will be aware of the limitations of these measurements.37
A difficult aspect of reliability is that the patient may have developed a different understanding of the pain problem and may give a different response from one examination to the next. It is equally important for the examiner to ask himself or herself whether the interpretation of the patient's responses differs from one examination to the next. Both factors affect the reliability of the information being gathered.37
Perhaps it is worthwhile to reexamine the concepts of subjective and objective measurements. Sometimes the terms "objective" and "subjective" are concerned not with the reliability of a measurement, but with the nature of what is being measured. It could be argued that pain is a subjective phenomenon, but if it is measured reliably, the quality of the measurement would be objective.
Acknowledgements
Philippe Harari and Karen Legge (2001), Psychology and Health, Heinemann, 0-435-80659-9. Highly recommended, easy to read, affordable text; a must have for every student.
LINKS
Continue with Controlling Pain
Return to Nature and Symptoms of Pain