"
The nocebo effect
Dr Sundararajan Rajagopal,
Consultant Psychiatrist
South London & Maudsley NHS Foundation Trust
Adamson Centre for Mental Health
St. Thomas' Hospital
London SE1 7EH
The nocebo effect
The wide range of responses that placebos elicit is a subject that has
intrigued researchers for many years. A placebo, in general, is an inert
substance that has no inherent pharmacological activity. It looks,
smells and tastes like the active drug with which it is being compared.
The term 'nocebo effect' refers to the negative consequences arising
from the administration of a placebo. The same placebo can result in
both a significant placebo effect such as improvement, and a prominent
nocebo effect such as a major adverse reaction. If a placebo causes
mainly nocebo effects, it would be called a nocebo; in research studies
an artificial procedure inducing an aversive symptom (e.g. causing arm
pain by increasing blood pressure using a tourniquet) is also known as a
nocebo. Walter Kennedy introduced the concept of 'nocebo reaction' in
1961, a few years after Henry Beecher published his landmark paper on
the placebo effect.1,2 Kennedy attributed it to a "quality inherent in
the patient, not the remedy". While the better studied placebo effect
has itself still not been fully explained, the nocebo effect is even
poorly understood.
The main psychological mechanisms for the nocebo effect, as for the
placebo effect, seem to be the related factors of subconscious
conditioning and conscious expectations. Patients who have had previous
experience of prominent side-effects with an active drug may be
conditioned to experience similar side-effects subsequently even when
they are only given a placebo. Patients may also anticipate adverse
effects if they are influenced by the knowledge of the list of potential
side-effects described by researchers before entering a
placebo-controlled trial, and also by the memory of past side-effects
suffered by oneself or others. Obviously, conditioning and expectations
would be prominent during placebo therapy only if the patient is blind
to the fact that they are now receiving an inert treatment.
Biological factors also appear to play a role. Nocebo hyperalgesia, an
increase in pain from administration of a non-nociceptive substance, has
been shown to be mediated by cholecystokinin, and to be abolished by the
cholecystokinin-antagonist proglumide.3 Another study, in which arm pain
was induced in healthy volunteers, suggested that nocebo hyperalgesia
was associated with hypothalamic pituitary adrenal (HPA) axis
hyperactivity; proglumide antagonised the hyperalgesia but not the HPA
axis hyperactivity.4 In a different experiment, volunteers had arm pain
induced and then were given an injection (saline) with some being told
that it was a pain relieving and others advised that it was a pain
increasing (nocebo) substance5; cortisol levels increased in both groups
but was greater in the nocebo group, suggesting that expectation of pain
increase was an added factor.
While the majority of studies on placebo and nocebo effects show results
consistent with patient expectations, it has also been suggested that
negative expectations may paradoxically have a protective effect. In one
study, motion sickness was induced by an optokinetic drum, prior to
which volunteers were given inert pills, some with the message that the
pill was anti-nauseous and those in the nocebo group being informed that
it would make nausea worse.6 Contrary to expectations, the latter group
developed significantly fewer symptoms.
A review of more than a hundred phase I double-blind, placebo-controlled
trials found an overall rate of spontaneously reported side-effects in
those receiving placebo of 19%, with higher rates following repeated
dosing and in the elderly7; the commonest adverse events were headache,
drowsiness and asthenia. A history of adverse reactions to drugs is
likely to increase the rate of nocebo effects in future; in a trial of
600 patients with such a past history, 27% developed side-effects, with
a significantly higher occurrence in females.8 Another study of phase I
healthy volunteers showed that those with Type A personality traits were
almost thrice as likely to report side-effects after placebo
administration as subjects with Type B personality features.9 A review
of antidepressant trials found a slightly higher rate of nocebo effects
in females.10
Placebo and nocebo effects are not just restricted to inert substances,
but are ubiquitous phenomena that follow active treatments too. Hence,
not all symptoms that develop after taking an active drug are
necessarily due to the drug. When surveyed, a significant proportion of
the general population report a range of physical symptoms,11 that are
similar to the common non-specific side-effects of drugs. Therefore, the
nocebo effect recorded systematically during a placebo-controlled trial
is likely to also include pre-existing and spontaneously occurring
symptoms that are independent of the placebo. In the same way, these
trials may also be overestimating the frequency of non-specific
side-effects of the active drugs being compared with placebos.
Just as a proportion of the positive outcomes of a successful drug may
be explained by the placebo effect, so may a proportion of the negative
consequences by the nocebo effect. In clinical practice, it is vital for
the physician to not immediately attribute all the side-effects,
especially the non-specific ones, following drug administration to the
drug; it is important to consider the possible role of a drug-unrelated
nocebo effect. Although the concepts of placebo and nocebo effects have
been in modern medicine for about half a century, our knowledge of these
fascinating fields is still far from complete, and so they certainly
warrant, and also offer scope for, further research.
"
No comments:
Post a Comment