Influence of pharmaceutical funding on the conclusions of meta-analyses
BMJ 2007; 335 doi: https://doi.org/10.1136/bmj.39381.655845.BE (Published 06 December 2007) Cite this as: BMJ 2007;335:1167
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I am responding to Augusto Pimazoni's posting in which he complains
about the confusion caused by those who believe that claims of
effectiveness and safety made for pharmaceuticals should be based on
evidence.
In his response, the author cites the controversy over the safety of
rosiglitazone and characterizes it as "sad" and with "catastrophic"
effects." He anchors this rather stunning condemnation, in part, on the
incorrect assertion that the conclusions of different meta-analysis were
"most confusing and unreliable."
I served as the consumer representative on the U.S. Food and Drug
Administrations panel that heard the evidence about the cardiovascular
safety of rosiglitazone (and was the lone vote to take it off the market).
The results of three separate meta-analysis were presented; one by the
sponsor, one by the FDA reviewers and of course, the one by Nissen and
colleague published in the New England Journal of Medicine. All three
arrived at more or less the same conclusion; that there was a suggestion
rosiglitazone might pose excess CV risk. However, when the placebo-
controlled studies were singled out, the evidence of CV risk turned out
to quite robust. These results were confirmed a short time later by
Furberg et al.
I fail to understand how the significant relative risk revealed in
the placebo studies can be characterized as "unreliable and most
confusing." What evidence that a drug is unsafe would satisfy Dr. Pimazoni
if such studies do not?
What is truly "sad" about the rosiglitazone controversy is that
clinician-specialists once again have shown their unwillingness to be
guided in their practice by evidence that a popular, heavily marketed drug
may be unsafe. Worse, if Dr. Pimazoni is any example, they want to shoot
the messenger.
Competing interests:
None declared
Competing interests: No competing interests
What is going wrong with clinical studies? What is going wrong with
our renowned scientists that are still considered the last line of defense
against the explosion of ignorance devastating the whole planet? What is
going wrong with us, practicing physicians, who are no longer capable of
correctly interpreting even the very basic clinical studies? What is
going wrong with meta-analyses and the messing conclusions it yields? Who
is to blame for the critical credibility crisis of medical sciences and
for the sense of insecurity and uncertainty that promotes more confusion
than conclusions? After all, what is going wrong with medical science as a
whole?
Statistics was once defined as the art and the science of proving
whatever you want to prove. No matter how absurd your thesis might be
there will always be a statistical method to help you prove your point. As
a medical marketing consultant for the pharmaceutical industry in Brazil
and Latin America, I once had the opportunity of producing in a same week
two sets of arguments, pro and against cox-2 inhibitors, with totally
conflicting conclusions but fully based on reliable articles and authors
from the international literature. I did so in a very ethical way, without
any guilt or ethical conflict, considering the scientific quality of the
sources I used. Obviously, it would be impossible to carry out clinical
studies without using statistics. But the extraordinary development of
this branch of science produced a paradoxical side effect of complicating
the planning and the interpretation of clinical studies to the extent of
bringing much more fight than light to the medical science.
Although considered level A in evidence based medicine, meta-analysis
is now facing its most intense phase of lost credibility. Take the
rosiglitazone affair as a sad example of the catastrophic effects of
misguided selection and data misinterpreting. Quite a few different meta-
analyses were conducted by different authors on this subject, leading to
the most confusing and unreliable conclusions. Chances are that we might
be crucifying an effective drug solely on the basis of statistical
disputes with no bearing to the sacred medical truth. The arguments pro
and against rosiglitazone are fully derived from conflicting results and
conclusions of confusing and unreliable utilization of meta-analyses.
AUGUSTO PIMAZONI, MD
Coordinator, Diabetes Education and Control Group of the Kidney and
Hypertension Hospital, Federal University of Sao Paulo (UNIFESP), Brazil
Medical Marketing Consultant, Sao Paulo, Brazil.
pimazoni@uol.com.br
Potential conflicts of interest: Medical marketing consultant for the
pharmaceutical industry, but not a consultant for GSK.
Competing interests:
Medical Marketing Consultant to the Pharmaceutical Industry in Brazil and Latin America
Competing interests: No competing interests
Beyond any reasonable doubt
This legal concept is fully applicable to scientific matters for the
very simple reason that medical science should be guided by dispassionate
and unbiased evaluation of available data and its implications are crucial
to establish medical truth with the needed level of reliability. The drug
approval and continuous evaluation processes must follow the principle
that facts should be considered “beyond any reasonable doubt” to provide
adequate grounds for the approval or withdrawal of a drug from the market.
In response to Arthur Levine I have the following comments:
1-) First of all, unfunded criticism and personal innuendos do not belong
to the healthy practice of democratic debate and is a factual
demonstration of passionate, inappropriate and biased attitude towards the
very basic ethical foundations that should guide the exchange of ideas and
opinions about medical matters. And this is undebatable…
2-) Many meta-analyses can indeed be methodologically flawed because of
threats such as combination of dissimilar studies, publication bias and
inclusion of poor-quality studies (1,2). And this is undebatable…
3-) If the debatable conclusions of these meta-analyses were as strong as
they intended to be, FDA, EMEA and other regulatory agencies around the
world would have instantly withdrawn rosiglitazone from de market, as they
did with many other drugs. To my knowledge this didn’t happen so far. And
this is undebatable…
To make a long story short, the concept of “beyond any reasonable doubt”
is probably the most basic requirement for a wise decision on the issues
surrounding the safety profile of rosiglitazone or any other drug: if
proven guilty, this debate will be over. And this is undebatable…
AUGUSTO PIMAZONI, MD
Coordinator, Diabetes Education and Control Group – Kidney and
Hypertension Hospital – Federal University of São Paulo, Brazil.
Competing interests: Medical Marketing Consultant for the
Pharmaceutical Industry in Brazil and Latin America (Not a consultant for
GlaxoSmithKline)
Reference:
(1) : “Applying a Meta-Analysis to Daily Clinical Practice”. Miser, WF,
In: Evidence-Based Clinical Practice – Concepts and Approaches – Editors:
Geyman, JP, Deyo, RA and Ramsey, SD. Butterworth Heinemann, 2000. Chapter
6, page 59.
(2) : “Misleading Meta-Analysis: A Need to Look Beyond the
Headlines”. Chaturvedi N. et al. Diabetic Medicine. DOI: 10.11 11/j.1464-
5491.2007.02139.x
Competing interests:
Medical marketing consultant for the pharmaceutical industry in Brazil and Latin America
Competing interests: No competing interests