Interactive theories

Page 312
Black blue
312 HEALTH PSYCHOLOGY
Interactive theories
It is therefore necessary to understand the process of placebo effects as an active process,
which involves patient, treatment and health professional variables. Placebo effects
should be conceptualized as a multi-dimensional process that depends on an interaction
between a multitude of different factors. To understand this multi-dimensional process,
research has looked at possible mechanisms of the placebo effect.
Experimenter bias
Experimenter bias refers to the impact that the experimenter’s expectations can have
on the outcome of a study. For example, if an experimenter was carrying out a study
to examine the effect of seeing an aggressive film on a child’s aggressive behaviour
(a classic social psychology study) the experimenter’s expectations may themselves be
responsible for changing the child’s behaviour (by their own interaction with the child),
not the film.
This phenomenon has been used to explain placebo effects. For example, Gracely
et al. (1985) examined the impact of doctors’ beliefs about the treatment on the patients’
experience of placebo-induced pain reduction. Subjects were allocated to one of three
conditions and were given either an analgesic (a painkiller), a placebo or naloxone (an
opiate antagonist, which increases the pain experience). The patients were therefore told
that this treatment would either reduce, have no effect or increase their pain. The doctors
giving the drugs were themselves allocated to one of two conditions. They either believed
that the patients would receive one of three of these substances (a chance of receiving a
pain killer), or that the patient would receive either a placebo or naloxone (no chance of
receiving a pain killer). Therefore, one group of doctors believed that there was a chance
that the patient would be given an analgesic and would show pain reduction, and the
other half of doctors believed that there was no chance that the patient would receive
some form of analgesia. In fact, all subjects were given a placebo. This study, therefore,
manipulated both the patients’ beliefs about the kind of treatment they had received and
the doctors’ beliefs about the kind of treatment they were administering.
The results showed that the subjects who were given the drug treatment by the
doctor who believed they had a chance to receive the analgesic, showed a decrease in
pain whereas the patients whose doctor believed that they had no chance of receiving
the pain killer showed no effect. This suggests that if the doctors believed that the
subjects may show pain reduction, this belief was communicated to the subjects who
actually reported pain reduction. However, if the doctors believed that the subjects
would not show pain reduction, this belief was also communicated to the subjects
who accordingly reported no change in their pain experience. This study highlights
a role for an interaction between the doctor and the patient and is similar to the
effect described as experimenter bias described within social psychology. Experimenter
bias suggests that the experimenter is capable of communicating their expectations
to the subjects who respond in accordance with these expectations. Therefore, if
applied to placebo effects, subjects show improvement because the health professionals
expect them to.
Page 312
Black blue
Page 313
Black blue
PLACEBOS AND THE INTERRELATIONSHIP BETWEEN BELIEFS, BEHAVIOUR AND HEALTH
313
Patient expectations
Research has also looked at the expectations of the patient. Ross and Olson (1981)
examined the effects of patients’ expectations on recovery following a placebo. They
suggested that most patients experience spontaneous recovery following illness as most
illnesses go through periods of spontaneous change and that patients attribute these
changes to the treatment. Therefore, even if the treatment is a placebo, any change will
be understood in terms of the effectiveness of this treatment. This suggests that because
patients want to get better and expect to get better, any changes that they experience are
attributed to the drugs they have taken. However, Park and Covi (1965) gave sugar pills
to a group of neurotic patients and actually told the patients that the pills were sugar
pills and would therefore have no effect. The results showed that the patients still showed
some reduction in their neuroticism. It could be argued that in this case, even though the
patients did not expect the treatment to work, they still responded to the placebo. However, it could also be argued that these patients would still have some expectations that
they would get better otherwise they would not have bothered to take the pills. Jensen
and Karoly (1991) also argue that patient motivation plays an important role in placebo
effects, and differentiate between patient motivation (the desire to experience a symptom
change) and patient expectation (a belief that a symptom change would occur). In a
laboratory study, they examined the relative effects of patient motivation and patient
expectation of placebo-induced changes in symptom perception following a ‘sedative
pill’. The results suggested a role for patient expectation but also suggested that higher
motivation was related to a greater placebo effect.
Reporting error
Reporting error has also been suggested as an explanation of placebo effects. In support
of previous theories that emphasize patient expectations, it has been argued that
patients expect to show improvement following medical intervention, want to please the
doctor and therefore show inaccurate reporting by suggesting that they are getting
better, even when their symptoms remain unchanged. (In fact the term ‘placebo’ is
derived from the Latin meaning ‘I will please’.) It has also been suggested that placebos
are a result of reporting error by the doctor. Doctors also wish to see an improvement
following their intervention, and may also show inaccurate measurement. The theory of
reporting error therefore explains placebo effects in terms of error, misrepresentation or
misattributions of symptom changes to placebo. However, there are problems with the
reporting error theory in that not all symptom changes reported by the patients or
reported by the doctor are positive. Several studies show that patients report negative
side effects to placebos, both in terms of subjective changes, such as drowsiness, nausea,
lack of concentration, and also objective changes such as sweating, vomiting and skin
rashes. All these factors would not be pleasing to the doctor and therefore do not
support the theory of reporting error as one of demand effects. In addition, there are
also objective changes to placebos in terms of heart rate and blood pressure, which
cannot be understood either in terms of the patient’s desire to please the doctor, or the
doctor’s desire to see a change.
Page 313
Black blue
Page 314
Black blue
314 HEALTH PSYCHOLOGY
Conditioning effects
Traditional conditioning theories have also been used to explain placebo effects (Wickramasekera 1980). It is suggested that patients associate certain factors with recovery and
an improvement in their symptoms. For example, the presence of doctors, white coats,
pills, injections and surgery are associated with improvement, recovery, and with effective
treatment. According to conditioning theory, the unconditioned stimulus (treatment)
would usually be associated with an unconditioned response (recovery). However, if this
unconditioned stimulus (treatment) is paired with a conditioned stimulus (e.g. hospital, a
white coat), the conditioned stimulus can itself elicit a conditioned response (recovery,
the placebo effect). The conditioned stimulus might be comprised of a number of factors,
including the appearance of the doctor, the environment, the actual site of the treatment
or simply taking a pill. This stimulus may then elicit placebo recovery. For example, people
often comment that they feel better as soon as they get into a doctor’s waiting room, that
their headache gets better before they have had time to digest a pill, that symptoms
disappear when a doctor appears. According to conditioning theory, these changes would
be examples of placebo recovery. Several reports provide support for conditioning theory.
For example, research suggests that taking a placebo drug is more effective in a hospital
setting when given by a doctor, than if taken at home given by someone who is not
associated with the medical profession. This suggests that placebo effects require an
interaction between the patient and their environment. In addition, placebo pain reduction is more effective with clinical and real pain than with experimentally created pain.
This suggests that experimentally created pain does not elicit the association with the
treatment environment, whereas the real pain has the effect of eliciting memories of
previous experiences of treatment, making it more responsive to placebo intervention.
Anxiety reduction
Placebos have also been explained in terms of anxiety reduction. Downing and Rickles
(1983) argued that placebos decrease anxiety, thus helping the patient to recover. In
particular, such a decrease in anxiety is effective in causing pain reduction (Sternbach
1978). For example, according to the gate control theory, anxiety reduction may close
the gate and reduce pain, whereas increased anxiety may open the gate and increase
pain (see Chapter 12). Placebos may decrease anxiety by empowering the individual and
encouraging them to feel that they are in control of their pain. This improved sense of
control, may lead to decreased anxiety, which itself reduces the pain experience. Placebos
may be particularly effective in chronic pain by breaking the anxiety–pain cycle (see
Chapter 12). The role of anxiety reduction is supported by reports that placebos are
more effective in reducing real pain than reducing experimental pain, perhaps because
real pain elicits a greater degree of anxiety, which can be alleviated by the placebo,
whereas experimentally induced pain does not make the individual anxious. However,
there are problems with the anxiety reducing theory of placebos. Primarily, there are
many other effects of placebos besides pain reduction. In addition, Butler and Steptoe
(1986) reported that although placebos increased lung function in asthmatics, this
increase was not related to anxiety.
Page 314
Black blue