Healthy Skepticism Library item: 5859

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.
Letter to the Editor: Taming Who?
BMJ 2006 Jul 29
http://bmj.bmjjournals.com/cgi/eletters/333/7558/0-f#138555

Full text:

BMJ
Taming Who?
29 July 2006
David T Healy,
Professor of Psychiatry
Cardiff University LL57 2PW
Dear Sir

After approving SSRI antidepressants in the late 1980s and early 1990s, on
the basis of slender evidence of efficacy, the Food and Drug Administration
(FDA) in the United States (US) asserted there was “no credible evidence”
that these drugs could cause violent and suicidal behaviour in some
vulnerable individuals – despite what the evidence available from the late
1980s indicated. No warning was issued for prescribers or consumers. The new
antidepressant market ballooned to a $20 billion dollar business.

The US is the biggest profit centre for one of the richest industries in the
world. But there remains for the drug companies a significant obstacle
towards even greater profits: product liability awards. When drug companies
fail to disclose important risks, US courts can award large sums, and in an
effort to keep the risks and their concealment hidden, settlement costs are
now a routine part of doing business on company ledger sheets.

At someone’s behest – and this is the key point germane to Fiona Godlee’s
question as to whether we can tame the monster (1) – FDA came to the aid of
the pharmaceutical industry by including in a new rule amending an existing
regulation regarding drug labelling, the argument that in the FDA’s view,
anyone suing a drug company for failing to disclose a risk should not be
allowed to do so because such a lawsuit is pre-empted by federal law (2).
The argument was that if a drug was sold, that meant it was approved by the
FDA, and that all risks that should be known are in fact known and
disclosed.

In selecting a test case for this legislation, the FDA placed itself in a
position of defending the drug companies against law-suits by people who
claimed they had been harmed by undisclosed risks of antidepressants. In so
doing, they have staked a position that it is illegal to warn patients about
a risk of suicide from antidepressants. This position has become
increasingly untenable as evidence accumulates so that even GlaxoSmithKline
now acknowledges the risk to be present (3). But FDA dug in and continues to
defend its claim even as its scientists protested the agency’s subjugation
of science and public health to political aims (4).

In 2003, the FDA first presented an analysis of suicide from clinical trials
of antidepressants, most of which had been completed a decade previously
(5). Analyses of suicides and suicide attempts in antidepressants trials had
been published previously by others, each showing that antidepressants
increased the risk of suicide. This result hid for years behind a
statistical smokescreen with the claim that the increased risk of suicide
with antidepressants should be disregarded because it was not
“statistically” significant (6). But the FDA, with a database of more than
40,000 patients in trials from all of the antidepressant manufacturers,
found an increased risk of suicide with antidepressants that was
“statistically significant.” Instead of concluding that their analysis
confirmed the increased risk, which would necessitate warnings on the drugs
and admit the fallacy of their pre- emption argument (currently being
defended in litigation with millions of dollars hanging in the balance), the
FDA concluded that with a few clever statistical adjustments, all of the
increased risk disappeared.

This FDA claim is dubious because in short term randomised drug trials large
amounts of bias sufficient to wipe out a doubling of risk are unprecedented
(6). Indeed, if these trials were so biased, then it would mean that the
studies that formed the scientific basis of FDA approval of these drugs were
seriously flawed. The data present the FDA with a dilemma: was the FDA
decision to approve antidepressants based on biased evidence, or were the
trials valid, in which case so is the “unadjusted” analysis which shows that
antidepressants cause suicide? On these data it would seem that FDA were
faced with the prospect of admitting to a huge mistake – either the mistake
of failing to warn against increased risk of suicidal behaviour or a mistake
in approving antidepressants in the first place. Fortunately for FDA, this
analysis only appeared as an abstract in a little-read journal and their
dilemma went unnoticed.

In the meantime, an FDA employee created a scandal by finding that
antidepressants increased suicidal behaviour in paediatric trials, and an
advisory committee recommended a black box, the highest level of warning,
indicating the increased risk for children to all antidepressants. And one
manufacturer, GlaxoSmithKline, admitted that trials in adults showed an
increased risk of suicidal behaviour too. But FDA continues on the path of
denial. This year, they released the “full” publication of their analysis of
suicidal behaviour in trials of antidepressants in adults (7). In their
latest release, which fails to mention or cite their previous analysis, the
FDA has claimed that the cases of suicide on the new antidepressants have
mysteriously vanished, and with them the previously reported increased risk.
In addition to all the unanswered questions raised previously, the FDA’s
latest story only raises more embarrassing questions. What happened to the
suicides? Why does the FDA now claim to be unaware of suicides that they
previously analysed and of a risk that even manufacturers have admitted?
(8).

The FDA is in a shambles, and there is considerable evidence that Britain’s
Medicines and Healthcare Regulatory Agency (MHRA) is not in any better shape
(9). The claims of these regulators in general are now no more credible than
their claims that Vioxx was safe and effective. Given that these are the
bodies that are supposed to tame the monster, the question posed by Fiona
Godlee can only at present be answered in the negative – unless the answer
is that our first step must be to radically reform these bodies. The key
questions otherwise are first who’s bidding precisely are the regulators now
following (are these politicians or businessmen or who are they?) and second
what can we do about that?

1/ Godlee F (2006). Editorial: Can we tame the monster? BMJ 333

2/ Dept. of Health and Human Serv., Docket No. 2000N-1269, “Requirement of
Content and Format of Labeling for Human Prescription Drug and Biological
Products,” p. 38 (Jan. 18, 2006).

3/ GlaxoSmithKline. Letter to healthcare professionals, May 2006.
www.gsk.com/media/paroxetine/adult_hcp_letter.pdf (accessed 13/05/2006).

4/ Union of Concerned Scientists survey:
http://www.ucsusa.org/scientific_integrity/interference/fda-scientist-
survey.html).

5/ Hammad TA, Mosholder A, Boehm G, Racoosin JA, Laughren T. Incidence of
suicides in randomized controlled trials of patients with major depressive
disorder. Pharmacoepidemiol Drug Safety 2003;12(suppl 1):S156.

6/ Healy D (2006). Did regulators fail over selective serotonin reuptake
inhibitors. BMJ 333, 92-95.

7/ Hammad TA, Laughren TP, Racoosin JA. Suicide rates in short term
randomised controlled trials of newer antidepressants. J Clinical
Psychopharmacology 2006; 26, 203-207.

8/ Healy D. bmjjournals.com/cgi/eletters/333/7558/92

9/ See correspondence on MHRA’s handling of trial suicide on
www.socialaudit.org.uk/6060116.htm#FOI

Competing interests: DH has links to all the major pharmaceutical companies
and has been an expert witness in legal actions involving antidepressants
and suicide