was not found to be effective above and beyond the placebo effect. However, from
a psychological perspective, the interesting question is, ‘Why is adherence to placebo
treatments associated with improved health outcomes?’
Placebo effects have raised doubts about a range of treatments. For example, although
the UK National Institute for Health and Clinical Excellence (2004) recommended
selective serotonin reuptake inhibitors (SSRIs such as Prozac) as the preferred treatment
for mild to severe depression, their effectiveness has been questioned because depressed
patients in placebo conditions show good levels of recovery (Kirsch and Sapirstein, 1998;
Quitkin, 2000). Placebo responses raise the possibility that observed treatment effec -
tiveness may not be due, or at least wholly due, to the pharmacological processes meant
to explain the operation of a medication but to psychological changes initiated by beliefs
that one is taking an effective treatment which, in turn, lead to health improvement
(Moncrieff and Kirsch, 2005).
In trials that only compare a treatment with a placebo condition, numerous factors
contribute to an observed ‘placebo effect’. Natural fluctuations in physiological
functioning mean that, for many conditions, some people spontaneously improve
(so called spontaneous remission). This is especially true of long-term conditions
when people are likely to seek medical help when their symptoms are most severe.
People involved in intervention trials may also change how they assess their symptoms
(cf. Norman and Parker, 1996). In addition, they may engage in new behaviours
relevant to their treatment. Controlling for such effects necessitates a three-condition
design in which treatment is compared to both placebo and a no-treatment control
group (e.g. people randomly assigned to a waiting list for trial inclusion). The true
placebo effect can then be defined as the additional gain seen in the placebo group
over and above the no-treatment condition (Ernst and Resch, 1995). Note too that
placebo conditions need to mimic the context in which treatment is administered.
These so-called ‘non-specific effects’ include setting and communication. This is why
health care professionals involved in trials should, ideally, be blind to which condition
a patient has been allocated to.
Explaining placebo effects: patient expectations and
anxiety reduction
Many explanations of placebo effects have been tested and it is likely that multiple
processes are involved (see Stewart-Williams, 2004, for a clear summary). For example,
placebo analgesic effects have been shown to be mediated by the release of endogenous
opioids (Levine, Gordon and Fields, 1978) demonstrating that some placebo effects
have measurable effects on the brain and endocrine system. However, such observations
do not explain howplacebo administration affects physiological systems. This may
operate through patient expectations and anxiety reduction. When a health care
professional (perceived to be competent) communicates to a patient that her problem
is understood and that she is being prescribed (or advised to undertake) an effective,
manageable treatment the patient will expect her condition to improve. Such
expectations may have cognitive, physiological and behavioural effects. For example,
believing that the worst is over could liberate the patient to devote time and energy
to other life-enhancing pursuits. It could alter coping strategies, perhaps leading to more
social support seeking. It is also likely to reduce stress and anxiety. Secondary appraisals
RELATING TO PATIENTS 245