Healthy Skepticism Library item: 12941

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: news

Healy D.
'Are clinicians trained to read the evidence? Sadly not'
The Guardian 2008 Feb 27

Full text:

Some people will be shocked to discover that Prozac has been prescribed so widely for decades when, in fact, it barely
works. However, the real story is even worse. First, the findings are not new, and it is not only the Prozac group of
antidepressants that we should be concerned about; second, the findings point to a general medical inability to
understand evidence; and, finally, they reveal the dark side of company marketing, and the role of regulators.
 
In 2006 – in a review that was much larger and more far-reaching than the recent one undertaken into Prozac – the US
Food and Drug Administration reviewed all antidepressant trials, with data from 100,000 patients. The FDA reported that
while five out of 10 people appeared to respond to the pills, four out of 10 responded to the placebo.
 
In a clinical trial, a drug is said to “work” when differences between the group taking it and the control group are
statistically significant. In sufficiently large trials even a small difference may be statistically significant. As a
result of this, drugs that are sedating or tranquillising can be deemed to “work for depression”. Provided the drug can
be shown to beat a placebo in two trials, the regulators are prepared to license it – even if it fails in the other 98
trials out of 100. The regulators acknowledge that what they do is take these hints that the drugs “might” work as
grounds to let companies market them as effective treatments.
 
And once regulators have approved such drugs, companies market them by selectively publishing from the trials
undertaken, in articles that are little more than marketing copy appearing under the apparent authorship of the biggest
names in the field and in the most distinguished journals.
 
But once marketed, surely clinicians are trained to read the evidence rather than just reach for their prescription pad?
Sadly not. If the four out of five responding to the placebo are stripped out of the apparent drug response, the figures
show only one out of 10 people have a true response to the antidepressant. Meanwhile, we know that the natural history
of depression means that many will improve within weeks whether treated or not. It is also thought that sensible advice
on matters of diet, lifestyle and alcohol intake, as well as basic problem-solving on work and relationship issues, may
make a difference. And it is suspected that our perceptions that we are being cared for by a medical expert may make a
difference, an effect that may be enhanced by being given a substance we think will restore some chemical balance to
normal – even if that imbalance is mythical.
 
Put this way it becomes clear that if clinicians are to follow the evidence, they should have a greater resort to
judicious waiting. As Philippe Pinel put it 200 years ago: “It is an art of no little importance to administer medicines
properly: but it is an art of much greater and more difficult acquisition to know when to suspend or altogether to omit
them.”
 
While many clinicians appreciate this point in the abstract, few practise medicine in this way or caution us that an
apparent response to treatment may not stem from the drug. This is not an argument for cognitive behavioural therapy
over pills. In fact, the evidence for CBT is similar to that for pills – the greater part of the response comes from
placebo factors.
 
The fact that the same arguments are made to “sell” both CBT and pills show that the problems lie deeper. Controlled
trials mandate interventions. And when the trials are conducted within a framework in which “No” is the one thing that
cannot be marketed, we risk being separated from the traditional art of medicine. What we gain in “facts”, we lose in
wisdom.