Gabriel szabo/guzelian
Interview
Aubrey Blumsohn
Academic who took on industry
Aubrey Blumsohn forfeited his job after going public with
concerns about access to Procter and Gamble’s research data
on the osteoporosis drug risedronate. Clare Dyer talks to him
about his experience
22
“Scientists since Galileo have realised you can’t be a
scientist without data,” observes Aubrey Blumsohn.
It seems a statement of the obvious, but he
welcomes the General Medical Council’s recognition
in the case of Richard Eastell, the former colleague
whom he reported to the GMC, that “data” mean
raw data, not summary data produced by a drug
company’s in-house statistician.1 2
That recognition, he believes, vindicates the
stand he took when he fought US based Procter and
Gamble (P&G) Pharmaceuticals, which refused him
access to the raw data for research Professor Eastell
and he were leading on the company’s osteoporosis
drug risedronate between 2002 and 2005. His
determination eventually forced the company to
release the data in 2006, but it cost him his job
as senior lecturer in metabolic bone medicine at
Sheffield University and led him to abandon his
career as a clinical researcher.
The GMC cleared Eastell, director of the bone
metabolism research unit at Sheffield University,
of dishonesty and misconduct but found that he
had failed to correct before publication an untrue
statement that all the authors in an earlier study he
led, published in the Journal of Bone and Mineral
Research in 2003, “had full access to the data and
analyses.”3 That research was also for P&G and on an
overlapping set of data from the same 1990s clinical
trials as Blumsohn’s work.
The GMC panel concluded that there was an
“evolving understanding of access to data” in 2002
when the statement was added to the research
paper by a P&G medical writer, but held that “data”
meant raw data and therefore that the assertion was
misleading. Although the panel said Eastell might
have been negligent, he had not acted dishonestly
and was not guilty of misconduct. The unit at
Sheffield, which he still heads, has won substantial
government funding for its research.
The charge faced by Eastell at the GMC related
only to his 2003 paper and not his joint research with
Blumsohn. Although it was Blumsohn who originally
reported Eastell to the GMC, he subsequently
withdrew as a complainant in protest at the glacial
pace of the investigation and what he described in a
letter to the GMC’s then president Graeme Catto as its
“apparent tolerance for indiscretions of highly placed
individuals.” Blumsohn insists nevertheless that
he wishes Eastell no ill, describing the professor, his
former PhD supervisor, as “an impressive academic;
a very bright guy.”
Academic and commercial relations
The saga of Sheffield and P&G highlights the tension
between commercial imperatives and scientific
integrity in a system in which researchers depend
heavily on drug companies to fund their work. P&G
was a major sponsor at Sheffield, where several
posts were funded by the company.
Blumsohn, now 48, was born and qualified as
BMJ | 2 january 2010 | Volume 340
Interview
a doctor in South Africa. He was among dozens of
doctors who left the country when their contracts
were not renewed after they refused to apologise
for a letter they wrote to the South African Medical
Journal protesting about conditions at Baragwanath
Hospital in Soweto.
After taking a PhD in Eastell’s unit, he worked in
Dundee for five years before returning to Sheffield,
where he held a senior lectureship and headed the
laboratory in Eastell’s department. After Eastell’s
initial work on risedronate, published in the Journal
of Bone and Mineral Research, Blumsohn and Eastell
signed a contract with P&G in mid-2002 to carry out
further measurements on blood and urine samples
stored from the clinical trials held during the 1990s.
The intention was to follow up Eastell’s paper with
two more papers. P&G sent Blumsohn two abstracts
based on its statistical reports, with Blumsohn
listed as first author. The company subsequently
submitted these to an international meeting. But
when Blumsohn requested information concerning
the randomisation codes showing who had taken
the drug, who took a placebo, who had fractures, and
who had not, the company refused to supply it, so he
was unable to check the reported findings. As a result
of the stand-off, the papers were never submitted for
journal publication.
At the time, risedronate was in fierce competition
with Merck’s osteoporosis drug alendronate, which
was thought to be a stronger drug in reducing bone
turnover and increasing bone density. Merck was
about to publish a head to head trial of the two
drugs. “Everyone knew it was going to show that
the Merck drug was a more powerful drug,” recalls
Blumsohn—though, ironically, he says, “I’m not sure
a more powerful drug is better.”
But the results of Eastell’s research, which
formed the basis of the 2003 paper,3 suggested that
risedronate’s lesser potency should not make it less
effective because there was a threshold beyond
which further reduction in urinary crosslinked
N-telopeptides of type I collagen (NTX), a key marker
of bone turnover, did not reduce the risk of fracture in
patients taking risedronate.
Requests by Blumsohn for the raw data were
repeatedly rebuffed by P&G, which claimed the data
belonged to the company and that it was standard
industry practice to limit access to raw data by
external researchers. Although P&G would not allow
Blumsohn to do an independent analysis, he was
invited to the company’s UK headquarters at Egham
in Surrey to look at the company’s analyses.
“I went down to Egham for a day, and it was at that
point that the whole thing fell apart. I wanted to see
the data plotted out on diagrams, so I could look and
see whether anything looks plausible. There was
this plot that showed immediately that everything
we’d been told was just nonsense. A substantial
proportion of patients taking risedronate fell off the
end of the graph.”
BMJ | 2 january 2010 | Volume 340 Maintaining integrity
Blumsohn’s story, documented in his Scientific
Misconduct blog,4 is cited as a cautionary tale about
what can happen to whistleblowers. But he doesn’t
see himself as a whistleblower and was annoyed
when the university, with which he raised the
problem, kept referring him to the head of human
resources. He thought it was a clear case of breach of
contract by P&G and that
Sheffield should act to
enforce its rights.
“The university said,
‘This is an issue of
research misconduct.’
I said, ‘It’s not an issue
of research misconduct,
it’s a company flouting
the terms of its contract
with the university.’ We
had a contract with the company; they refused to give
the data. From the point of view of the university there
was no more problem than that —it couldn’t have
been made more obvious to them.”
He engaged lawyers himself to write letters asking
for the data, to no avail. Losing patience that anything
would be done through official channels, he gave
the university notice in mid-summer 2005 that he
was going to the press, and contacted the Lancet, the
BBC, and the Times Higher Education Supplement.
He was suspended from the university and
found himself unemployed for six months. The
university presented him with a draft agreement
with an extensive gagging clause it wanted him to
sign as part of a severance package with a pay-off
of £145 000; he refused and took it to the press. He
won’t discuss the final terms on which he left, but
the Times Higher Education Supplement says he
accepted a six figure sum.
Along with the university job, he lost his NHS post
at Sheffield Teaching Hospitals NHS Trust. “I couldn’t
resign from just my university job; I had to resign from
everything.” A two day a week NHS job in pathology
was salvaged for him at Sheffield, where he survives
on tenuous fixed term contracts. He didn’t want to
leave Sheffield, where his children were in school
and his wife works as a staff grade paediatrician.
In 2006, after the media publicity, the company
finally released the data to the researchers and
produced a “bill of rights” for scientists. After
independent analyses of the data, a letter from
Eastell and his coauthors to the Journal of Bone
and Mineral Research acknowledged “some errors
and some poor practice” in the study.5 The journal
published an editorial stating that both extremes of
the original graphs had been cropped, and that the
reanalyses did not show a level below which further
reduction of NTX was not associated with greater
reduction in the incidence of fracture.6
P&G told the BMJ it regretted the
“misunderstanding” with Blumsohn over the data.
“At the time, it was common practice in the industry
to provide access to data to outside researchers
through a company statistician. Since then, however,
we initiated a researcher bill of rights, which explicitly
states that all researchers are given direct access to
all data, upon request.”
Eastell said: “The conclusion of the 2003 paper
was that the evidence for a non-linear relationship
between fracture risk
and bone resorption
markers was present. That
conclusion was confirmed
for CTX [crosslinked C
telopeptide, another
marker of bone resorption]
but not for NTX after the
reanalysis which was
published in 2007.
“The GMC recognised
that there was never any intention on my part when
I wrote the paper along with others in 2002 to
deliberately mislead about our access to the raw data
used in the study. Following concerns raised in 2005
over the paper, I took every step I could to address
these issues as quickly as possible.
“I have spent the last 30 years conducting
research into the cause, diagnosis, and treatment of
osteoporosis, publishing around 295 papers. I am
committed to continuing with medical research and
hope that this work will hasten the development of
better treatments for patients with osteoporosis.”
Would Blumsohn have done anything differently if
he had his time back? “I might not have done it in the
same way. But I have no regrets that I prevented the
two further proposed papers from being published.
“It’s hard to encourage anyone to speak out
about poor practice in the current environment.
This case sums up what has gone wrong with
systems set in place to ensure safety and integrity
in scientific medicine. It would help if regulators
put as much effort into responding to serious critics
and whistleblowers as they do producing glossy
brochures and yet more guidance.”
Clare Dyer is legal correspondent, BMJ [email protected]
“This contract was a model
contract. It said the academics
would be able to interpret the
data and write a report, and
there would be no restrictions
on publication”x
Competing interests: None declared.
1
Dyer C. GMC clears research dean of dishonesty. BMJ
2009;339:b4167.
2 GMC. Fitness to practise panel 2-5, November 2009, www.
gmc-uk.org/static/documents/content/Anon_Eastell_
Minutes.pdf.
3 Eastell R, Barton I, Hannon RA, Chines A, Garnero P, Delmas
PD. Relationship of early changes in bone resorption to the
reduction in fracture risk with risedronate. J Bone Miner Res
2003;18:1051-6.
4 Blumsohn A. Scientific misconduct blog. www.scientificmisconduct.blogspot.com/.
5 Eastell R, Hannon RA, Garnero P, Campbell MJ, Delmas
PD. Relationship of early changes in bone resorption to
the reduction in fracture risk with risedronate: review of
statistical analysis. J Bone Miner Res 2007;22:1656-60.
6 Eisman JA, Lorenzo JA. Challenges in science and academic
–industry interactions. J Bone Miner Res 2007;22:1654-5.
Cite this as: BMJ 2009;339:b5293
ЖЖFrom the archive: For more on whistleblowing, see
“The price of silence,” BMJ 2009;339;b3202
23
Medicine in the News
Health stories of the decade
Jeremy Laurance picks out the most important stories in the UK media over the past 10 years
Any selection of the top 10 health stories from the
past decade is bound to be a matter of dispute.
On some we can perhaps all agree, at least from
the UK perspective. The MMR (measles, mumps,
and rubella) vaccine, hospital infections,
National Health Service funding, and pandemic
flu have dominated UK headlines not just for
months but for years. Obesity, the smoking ban,
cancer drugs, and in vitro fertilisation have also
hogged their share of the limelight. The general
practitioner Harold Shipman demands inclusion
as, possibly, medicine’s nadir, and alcohol is in
there because there is a head of steam building
up behind it which makes it the likely focus of the
next big public health battle.
But what is striking, looking over these 10
s­tories, is how many of them—fully half—are in
the arena of public health, which is still a Cinderella specialty in medicine. Public health deals
in large numbers and scary themes—plague,
catastrophe—which lends itself to the news
agenda. But public health is also an area where
the citizen has a role. We choose what to eat and
drink, what vaccinations to have, what lifestyle to
follow. News media can influence those choices
and thus play their part, for good or ill, in the
health of the nation.
Pandemic flu
MMR vaccine
No vaccine in recent history has provoked so
much anger, fear, and ill informed speculation.
It started in 1998 with the publication of the
now infamous Lancet paper linking MMR vaccine with bowel disease and autism. Vaccination
rates with MMR stood at 91% in 1997-8 but had
slipped to 80% in 2003-4 and as low as 60% in
parts of London. Although the rates have since
recovered to 85%, hundreds of thousands of
children remain unprotected from the diseases
and cases of measles have soared.
One of the greatest puzzles of the saga is what
has sustained this level of mistrust in the medical authority. Unlike most scientific controversies, which flare up and die away, this one has
simmered for a decade. And it looks set to be
fired up again by the conclusion of the General
Medical Council case against the chief author of
the Lancet paper, Andrew Wakefield, which is
expected to conclude early in 2010.
Hospital acquired infections
Few would have guessed at the beginning of the
decade that meticillin resistant Staphyloccus
aureus (MRSA) and Clostridium difficile would
become household terms. That they have done
so can be attributed to one fact. When patients
are admitted to hospital they accept there is a risk
from the medical procedures they are to undergo.
But they do not expect to contract a new illness
while they are there.
Doctors in the past have seen hospital infections as an unfortunate but inevitable complication of medical care, to be dealt with by
administering large doses of antibiotics. In the
case of MRSA, more and more powerful antibiotics have been required, and in the case of C difficile, antibiotics have themselves been a cause
of the illness.
Responding to public concern, the government made cutting hospital infections a key priority and the NHS responded. Deaths from MRSA
TEK IMAGE/SPL
If there is one story that has dominated the decade, this is it. It started in 1997, with the outbreak of avian flu in Hong Kong that led to the
death of a 3 year old boy and the slaughter of one
million chickens. For a few days panic gripped
the city as virologists warned Armageddon was
at hand.
Six years later in 2003, a Chinese professor
of respiratory medicine visiting Hong Kong
sneezed in a lift at the Metropole Hotel and
sparked a global panic. The mystery virus he
carried—severe acute respiratory syndrome
(SARS)—spread to seven other guests staying
on the ninth floor of the hotel. Six weeks later it
had infected 2300 people in 17 countries and
claimed 80 lives.
SARS wasn’t flu but it was awfully like it.
Within months, as suddenly as it appeared, it
disappeared. By June 2003, it was over—but not
before the virus had infected 8000 people and
claimed 800 lives worldwide.
The following year, 2004, avian flu reemerged in poultry flocks in the Far East and
began its march across the globe. With a 60%
death rate in humans—it has so far infected 442
people and claimed 262 lives—it posed a serious potential threat.
It has not yet mutated into a form readily
transmissible between humans, as virologists
feared. Instead, out of the left field, came swine
flu from Mexico, to cause the first pandemic of
the 21st century. Fear of avian flu shaped the
world’s response to swine flu. So far it seems to
be a kitten not a tiger, but public health specialists are keeping their fingers crossed.
24
BMJ | 2 january 2010 | Volume 340
Medicine in the News
and C difficile are down by more than a quarter
from their respective peaks.
The NHS
In the five years from 2002 to 2007, the NHS saw
the biggest funding boost in its history with a
£43bn increase, equivalent to 50% in real terms.
Did it feel 50% better? That is a matter for dispute. There are more doctors and nurses, more
intensive care beds and scanners, better hospitals, shorter waiting times, fewer heart deaths,
improved cancer treatment, and better mental
health care.
The downside is that the NHS has got less efficient. And it is facing a bigger burden from the
lifestyle problems of obesity, poor diet, and lack of
exercise. Though the extra cash has bought extra
services, we have not had a big enough bang for
the bucks. Unless the NHS can improve productivity, the outlook for the tough post-credit crunch
years is bleak.
Obesity
We are too fat and getting fatter. It has been a constant theme of the past decade. We beat ourselves
up about it, but we don’t know what to do about
it. Acres of newsprint and hours of broadcasting
are devoted to different ways of dieting, but as
soon as the weight is lost, most people put it back
on. Obesity is seen as a vanity issue not a health
issue. Only 6% of people in England understand
that it increases the likelihood of cancer, heart
disease, and diabetes and reduces life expectancy
by an average of nine years. It will be a theme of
the next decade too.
Smoking
out from a measure that he suspected would be
deeply unpopular in the bars of his hometown.
He eventually relented, and the ban was introduced with barely a murmur of protest, making
indoor air cleaner and healthier and, more importantly, changing attitudes to smoking—now the
preserve of a committed minority.
Cancer drugs
The past decade has been hailed as the golden
age for discovery of cancer drugs with a slew
of new products on the market. None is better
known than trastuzumab (Herceptin). When
the breast cancer drug was licensed for early
stage disease in Britain in 2006 it aroused
huge excitement among researchers, patients,
and the public—principally because there had
previously been so little progress in developing effective cancer treatments.
But the new cancer drugs are posing a big
headache for the NHS because of their cost.
Drug companies have priced many new cancer
drugs too high for the benefits they bring, and
the National Institute for Health and Clinical
Excellence has rejected them.
The government has been forced to introduce special rules to allow patients to buy the
drugs privately while continuing to be treated
on the NHS. This is a problem that can only
get worse.
In vitro fertilisation
Nothing excites the public—and hence the
media—more than advances in reproductive
medicine. Ever since the birth of Louise Brown,
the world’s first test tube baby, 31 years ago the
idea of creating human life in the laboratory has
exerted a unique fascination.
Last summer scientists claimed they had created human sperm from stem cells derived from
a 5 day old male embryo. It raised the possibility
of a limitless supply of sperm made from a single stem cell line, after which there would be no
further reproductive need for men. Even though
AJ PHOTo/SPL
The ban on smoking in public places, introduced
in England in July 2007 and earlier in the other
UK countries, was Liam Donaldson’s finest hour.
The chief medical officer first made the proposal
in 2003 and resurrected it in 2004, to the fury of
the then health secretary John Reid, a Glaswegian former smoker who feared the political fall
For the latest information from the BMJ
Group about pandemic flu,
visit http://pandemicflu.bmj.com/
BMJ | 2 january 2010 | Volume 340 the claims were challenged by other experts, it
was an irresistible story that guaranteed the
headlines that duly followed.
Harold Shipman
The decade opened with the jailing on 31 January 2000 of the most infamous doctor in history,
found guilty of murdering 15 of his patients and
suspected of having murdered 150 more. Six
years later, the government announced the biggest shake up of medical regulation in 150 years,
including reform of the GMC, five yearly checks
on doctors’ performance, and the appointment
of medical inspectors in every NHS trust.
Although the changes were already in train,
the Shipman case ensured that they were driven
through. Last month the GMC announced that
all doctors must now have a licence to practise,
as well as being on the medical register, which
will be conditional on successfully passing the
test for revalidation from 2011. Ultimately,
therefore, Britain’s worst serial killer may in
part be credited with making medicine safer for
the public.
Alcohol
The tobacco war has been won; now it is the
turn of alcohol. Where smokers were once targeted, drinkers, especially binge drinkers, now
find themselves in the spotlight. The British
Medical Association, the Royal College of Physicians, and Liam Donaldson are among those
who have joined the lobby for tough action to
curb Britain’s drinking culture. A minimum
price for alcohol, restrictions on promotions,
and tougher rules on advertising are among the
demands. Alcohol and its effects—for better or
ill—have dominated the press and the airwaves
as 2009 draws to a close. It will continue to do
so in 2010 and beyond.
Jeremy Laurance health editor, Independent
[email protected]
Competing interests: None declared.
Cite this as: BMJ 2009;339:b5281
REX
Jeremy Laurance also presents his review of the
stories of the decade in a BMJ podcast.
Listen at http://podcasts.bmj.com/bmj/
25