Wednesday, November 04, 2009

PharmaGossip wishes Aubrey Blumsohn all the best


Clare Dyer writes in the BMJ:

The former research dean of Sheffield University’s medical school was accused of dishonesty at a General Medical Council hearing on 2 November for allowing a journal to publish a false claim that he had seen all the data in a research study of which he was the lead author.

Richard Eastell (pic), who heads the bone research unit at Sheffield University, is at the centre of a research ethics case that saw the other main investigator, Aubrey Blumsohn, who raised questions over the data, take a financial settlement to leave his job at the university.

Dr Blumsohn says that he and Professor Eastell both asked Procter & Gamble, which was funding the study into its osteoarthritis drug risedronate (Actonel), for access to the full data but were refused. The university had been asked to carry out measurements on blood and urine samples that had been taken during clinical trials in the 1990s for a study comparing the bone density measurements of women who had and had not taken the drug.

The results were sent to Proctor & Gamble, which carried out analyses and sent them to the research unit for interpretation. Dr Blumsohn wanted to carry out his own analyses and asked for the codes showing which women had taken the drug, which had taken a placebo, and which had fractures, but the company refused.

The GMC alleges that Professor Eastell’s fitness to practise is impaired because of his misconduct. But at the start of the hearing the fitness to practise panel agreed to allow the charges to be amended, watering down their seriousness somewhat.

Professor Eastell was originally charged with having personally told the Journal of Bone and Mineral Research, which published the article in 2003 (18:1051-6; doi:10.1359/jbmr.2003.18.6.1051), that "all authors had full access to the data and analyses." The amended charge now says only that the article made that statement.

He is now accused of dishonesty not in making the statement but in not correcting it before it was published in the journal.

In 2006 Proctor & Gamble finally released the raw data and an independent analysis was carried out. In 2007 the Journal of Bone and Mineral Research carried a letter from Professor Eastell and the other authors admitting that only one author, the statistician from Procter & Gamble, had seen all the data when the article was published (2007;22:1656-60, doi:10.1359/jbmr.07090b).

The letter also acknowledged that the independent analysis "identified some errors and some poor practice" in the original paper. Among other errors, the authors were unaware that a crucial graph had been cropped in an asymmetrical way and excluded "between 34% and 49% of the more extreme values."

The hearing is scheduled to finish on 9 November.

More about Aubrey

No comments: