Sunday December 4, 2005
The Observer

Every medicine prescribed in Britain is offered to patients only after years collecting data from clinical trials to show a drug's safety and effectiveness. Increasingly, medical research in the UK is a collaborative effort between big pharmaceutical companies and universities. Although there will be a commercial contract between the two, much of that relationship is built on trust.

At its heart is the practice of 'ghost writing', pharmaceutical companies hiring writers to produce reports which academics give permission for their name to be put to.

It is rare for tensions in the highly sensitive world of scientific drugs research to break into the open. But an Observer investigation has uncovered a story of a war between one of Britain's most experienced academics and one of the world's most successful drugs companies. At issue is a highly significant report on a drug to treat a debilitating disease which the academic says was published under his name when he had not seen analysis of the research it was based on. The drugs firm denies anything untoward has happened.

In the summer of 2002, Dr Aubrey Blumsohn, a senior lecturer and honorary consultant in metabolic bone disease at Sheffield University, was preparing to lead a major research project. One of the world's largest pharmaceutical firms, Procter and Gamble, had just signed a £180,000 deal for his research team to carry out a study on one of their drugs, Actonel.

Actonel is a highly successful medication prescribed by doctors to nearly one million women around the world every year to treat osteoporosis. Also known as risedronate, it has been licensed as a treatment in the UK since 2002, and is known to be safe, with a good record of reducing the risk of broken hips, spines and wrists.

Blumsohn, who was born in South Africa, joined the Sheffield centre for bone metabolism research five years ago. He was to lead a third thorough investigation into the drug to look at how it was affecting the bones of thousands of women. To do the study, thousands of blood and urine samples, taken from women in more than 100 research centres around the world, were analysed in Blumsohn's laboratory in September 2002.

More than 3,000 pages of laboratory data and electronic records had to be put together by the researchers. Their aim was to look at the samples collected over a period of three years from women who were on Actonel, to determine which patients suffered fractures, and match that to changes in their bone turnover and bone density with their treatment. Bone turnover, the rate at which the body replaces its own bone material, is an important factor in osteoporosis.

The data was sent off in December 2002 to be analysed by statisticians at Procter and Gamble's headquarters in Egham, Surrey. Blumsohn's colleague, Richard Eastell, professor of bone metabolism at Sheffield, had worked on previous collaborations with the company, and had published data prepared by them.

A great deal was at stake for Procter and Gamble, a US company with headquarters in Cincinnati, Ohio. Its fortune was built on soap and candles made by the Englishman, William Procter, and the Irishman, James Gamble, more than 170 years ago. With shrewd investment and its ownership of leading brands in washing powders and disposable nappies, it has become a major player in pharmaceuticals.

Actonel is one of P&G's 16 'billion-dollar brands', making the company between $500m and $1bn each year. But there is intense competition with its main rival, the drug Fosamax, manufactured by another pharmaceuticals giant, New Jersey-based Merck.

There has been much debate over how these drugs actually work. It is known that they do two things: increase the mineral content of the bone, known as bone mineral density (BMD), and reduce the rate of bone turnover. Bone is a living tissue which constantly regenerates itself, by losing and gaining calcium and protein. Bones are always developing tiny holes which are filled in as new material grows, but as women grow older the holes are not filled in properly and the bones are weakened. Both Fosamax and Actonel are 'anti-resorptive agents' - they prevent the bones from losing material too fast.

Blumsohn asked at an early stage if he could see the way in which the analysis of his research was being carried out by P&G. The contract he and Eastell had signed with the company stipulated that their final written report would include all data, and any interpretation necessary for analysis.

Although the two experts were providing the raw data from their laboratory, the numbers were meaningless in themselves because they were coded. As in all cases of this type of work, the researcher is 'blind' to the details of the data because they might be 'influenced' by that knowledge. The results cannot be understood until they are decoded - and P&G possessed the key.

It is standard practice for authors to ask to see details of the work that is prepared in their name. In fact, medical ethicists would argue that this should happen as a matter of course. It is, after all, a doctor's career and reputation which is at stake.

Eastell, who was on the company's UK scientific advisory board, wrote on Blumsohn's behalf to Ian Barton, a senior statistician at P&G, asking if Blumsohn could see how the analysis was being carried out. In an email, Eastell, a highly respected expert in the field, made it clear he wanted to avoid criticism because questions had already been raised about how much of the data on the drug's performance he had seen in the past.

But to his and Blumsohn's surprise, their request was turned down. Barton replied by email: 'No, we do not intend for someone else to do the analysis. Resource wise, it wouldn't make sense for me to spend a lot of time training someone else to do the analyses.'

This was backed up by an email from a senior P&G official, Mike Manhart, director of clinical development at the company's headquarters in Cincinnati. He wrote to Eastell: 'I think we should look carefully at the pros and cons of Dr Blumsohn conducting the analysis you refer to.

'On the plus side it does add an extra layer of external "credibility". With this however, Industry [sic] loses the opportunity to demonstrate its ability to be a true partner in scientific endeavours. Beyond this, the practical issues to training up a new statistician and the corresponding delay in "time to result" may make the option difficult.'

Over the next nine months, Blumsohn made several requests to see data, all of which were turned down.

By April 2003, Blumsohn was becoming more worried. P&G staff had produced their own analyses and submitted two research papers to be presented at the American Society for Bone and Mineral Research conference in Minneapolis later that year, with him named as the main author. This was the main conference for experts on bone metabolism - where he would be grilled by his peers about the results. Yet he had still not seen any analysis of data.

The doctor discovered in July that a third paper had been written in his name, which he was expected to present at the conference. He only discovered this because someone from P&G wrote to congratulate him on his success.

The pressure was also on to submit papers for publication in journals. The company told Blumsohn that they had an 'external medical writer', Mary Royer, who could help to write it. These 'ghostwriters' are employed by companies to draft articles which are then submitted in the name of academics who have given their permission.

Barton emailed Blumsohn on 24 April 2003 to tell him: 'Mary is based in New York and is very familiar with both the risedronate data and our key messages, in addition to being well clued up on competitor and general osteoporosis publications.'

There has been concern about the use of such writers. The Observer revealed two years ago how hundreds of articles in medical journals claiming to be written by academics or doctors have been written by ghostwriters who were paid by drug companies. Papers in journals backing certain drugs have huge influence on which drugs doctors prescribe and the treatment hospitals provide. While doctors who have put their names to the papers can be paid well for doing so, the ghostwriters remain hidden. However, all medical journals subscribe to a code of practice which states that authors must be given full access to data and the interpretation of it.

In another email seen by The Observer, P&G say they want to achieve a 'paradigm shift' in doctors' attitude to Actonel. Previous research published by Eastell concluded that there was a 'plateau effect' with treatment, meaning that it produces benefits but only up to a certain level of depleted bone density.

One of P&G's rivals, Merck & Co, manufactures Fosamax, the market leader with sales of nearly $3bn a year worldwide. It was known to be better at increasing bone density and reducing bone turnover than Actonel. P&G wanted to be able to prove that once a drug had suppressed the reduced bone turnover by a certain amount - around 30 to 40 per cent - any further fall in turnover, which Fosamax could achieve, would not result in fewer fractures.

After numerous approaches P&G finally agreed that Blumsohn could see limited data at the end of July. He made the trip to P&G's Egham headquarters and Barton showed him the key graphs on his computer. Blumsohn was convinced that a different interpretation could be put on the figures than the one provided by P&G.

He was particularly concerned that around 40 per cent of the data appeared to have been left off the bottom of the graph. Blumsohn believed that had the data been included it would have disproved the plateau effect. As it stood, the graph appeared to support the existence of such an effect. The drugs company has denied that the data were in any way inaccurate or 'skewed' to get to the result they wanted.

In a statement to The Observer, the company said: 'For post hoc exploratory analyses, it is standard industry practice to limit access to the raw data by external researchers. Typically, analyses developed by or in collaboration with external researchers will be performed by company statisticians, and the results shared with the researcher.

'Occasionally, the researcher is given temporary and limited access to the data, to perform the analyses directly. Our policy is to allow external researchers sufficient access to the data and to perform those analyses necessary for them to be confident and comfortable with the conclusions they state in scientific communications.'

There are senior medical figures who feel this approach is inherently flawed. Stephen Evans, professor of pharmacoepidemiology at the London School of Hygiene and Tropical Medicine, said that a company should allow an independent statistician an opportunity to re-run analyses if this is requested by academic partners.

'If a research partnership is a genuine one then there has to be complete openness. I don't know how you can put your name to a paper otherwise,' he told the Times Higher Education Supplement recently. In a further statement, P&G refuted suggestions that they had not been fully co-operative: 'The raw clinical data in question were generated by P&G at great effort and expense,' the company said. 'They are from the pivotal clinical trials that supported the approval of risedronate for osteoporosis in many countries. P&G has a very legitimate proprietary interest in these clinical trial data.

'Dr Blumsohn had full access to the raw clinical data relevant to the abstracts, posters and presentations published or made in his name and could (and did) request many additional analyses to make sure he was fully comfortable with the approach and conclusions.'

In September 2003 Blumsohn told Eastell, a research dean at the university, that he thought the plotted graphs were misleading. But Eastell, whose work for the university has attracted research grants from P&G of £1.6m in recent years, told him that they 'really had to watch it' with P&G.

Despite Eastell's attempts to get access to the data, he had to tread carefully. In a conversation which Blumsohn taped, Eastell said: 'The only thing that we have to watch all the time is our relationship with P&G. Because we are... we have the big Sheffield Centre Grant which is a good source of income, we have got to really watch it. So, the reason why I worry is the network within P&G is like lightning. So if Ian [Barton] is unhappy it goes to Arkadi [Chines, global medical director of P&G Pharmaceuticals] and before we know it, there is an issue, there is a problem.' In December, Eastell again warned him that they would be in 'even deeper' financial trouble without the funding from P&G. In November 2003, P&G had in fact amended its publication brief to include the 40 per cent left off the original graph.

Blumsohn was still unhappy and in May 2004 Blumsohn put in a formal complaint to Eastell. His main point contained in a letter was that 'no self-respecting scientist could ever be expected to publish findings based on data to which they do not have free and full access'.

Blumsohn then attempted to raise his concerns with the dean of the university, Professor Tony Weetman, and the vice chancellor, Professor Bob Boucher. The matter was passed onto the head of human resources for investigation. In June this year, Blumsohn wrote to her explaining the saga himself. A month later, after telling the university he was going to talk to medical journalists about the affair, he received a threat of disciplinary action. In September he was suspended on the grounds that 'his conduct over these past months amounts to and constitutes conduct that is quite incompatible with the duties of office'.

When The Observer asked the university for their response to these allegations they rebutted the suggestion that they ignored his concerns about the data. 'In a meeting in May 2004, the Head of School advised Dr Blumsohn to raise his concerns [about research conduct] through the proper channels in order for them to be investigated fully in accordance with the university's policy. No formal complaint was raised.'

They added: 'Since then, there has been almost continual communication from the university to Dr Blumsohn urging him to bring forward his concerns and evidence through the proper channels, in order that they may be fully and thoroughly investigated.

'The University of Sheffield has offered Dr Blumsohn a range of procedures, including the offer of an independent investigatory panel, and direct access to the head of the university's Research Office and other senior staff. Despite this, the University of Sheffield has not received any substantive complaint from Dr Blumsohn related to the Bone Metabolism Unit's work with the pharmaceutical company and thus the university has not been able to investigate this matter.' Professor Eastell has also expressed concerns that Blumsohn's claims are 'misleading' and says he would welcome the opportunity to give evidence to the correct channels.

This weekend, Blumsohn remains suspended, at home with his wife and children in Sheffield. He sees himself as a whistleblower, revealing tensions at the heart of the scientific and pharmaceutical world. This week his allegations will be discussed in parliament as part of a debate on the pharmaceutical industry. The concerns he has raised about the way in which medical research is conducted are likely to provoke a very heated debate.

· Additional reporting by Vivienne Parry