In early 2006, GAP client and Sheffield University senior faculty member Dr. Aubrey Blumsohn traveled to Washington, D.C. as part of a week-long effort to inform Congressional officials and journalists about data concealment and manipulation performed on behalf of Procter & Gamble (P&G) regarding that company’s osteoporosis drug, Actonel. The popular medication was primarily used by post-menopausal women to strengthen bones and help prevent fractures.
Blumsohn told a story that worsens the image of a major pharmaceutical, and raises questions about the overall state of effectiveness of the FDA in monitoring the drug approval process. Blumsohn, the intended lead author on a P&G commissioned study, exposed his belief that the data actually disproved P&G’s therapeutic claims for Actonel, stating the company used his name in medical reports despite his not having full access to data sets – information that he repeatedly asked for.
Blumsohn works for Sheffield University in England, where he was a senior medical faculty member specializing in bone health. In 2002, Sheffield entered into a contract with P&G to collect Actonel data – the purpose of which was to determine how the drug prevented bone fractures, and how this related to change in bone resorption (the rate at which bone is removed) and bone mineral density. Consistent with research protocols in collecting data, Blumsohn was blind to which research subjects received Actonel and who got placebos.
To later analyze the data, he needed its key, P&G’s “randomization codes.” Despite his repeated requests over 18 months, P&G denied him access to the data, even as it published ghost-written abstracts in his name falsely implying the therapeutic equivalence of Actonel to Merck’s Fosamax, the industry leader.
Researchers are supposed to be granted access to full data sets to reach informed conclusions. In 2001, an international coalition of the world’s leading medical journal editors called for all journal article authors to pledge that they had “full access” to drug study data. This was recommended in order to avoid data manipulation by corporate sponsors of product information (that editorial appears in the September 18, 2001 edition of the Annals of Internal Medicine).
Just a Glance
In 2004, P&G allowed Blumsohn to review, at one of its British facilities, what the company purported to be the actual data set. In reviewing the data, Blumsohn realized that numerous graphs (illustrating Actonel’s effectiveness in preventing bone fractures) omitted 40 percent of one data set, apparently manipulating results to suit P&G’s marketing objectives. P&G officials told Blumsohn that if these additional data were included in the results, the study would have favored a competitor’s drug – Merck’s Fosamax.
After that time, Blumsohn kept raising concerns regarding P&G with Sheffield University, which subsequently offered him a large sum of money to leave his post. He refused, because the settlement agreement required his silence. He was suspended by the university for speaking out about the issue to the BBC. GAP accepted him as a client after that time, to educate American audiences on the lack of effective of the drug, and serious lapses in the drug-approval process.
Proctor & Gamble’s Attempts to Spin the Story
In the months following Blumsohn’s visit to America, P&G continued to subvert the integrity of research regarding Actonel by suppressing the critical data even while touting its “new openness” in response to the scandal. Faced with growing public scrutiny in both England and the U.S., P&G launched a public relations offensive, announcing a “Bill of Rights” for researchers, stating that it would release the data from disputed clinical studies on Actonel to an “independent statistician” named by Sheffield University – the very university that suspended Blumsohn for speaking out about public safety.
Welcoming what he terms the company’s “battlefield conversion in issuing the ‘Bill of Rights,’” Blumsohn maintained that “Procter and Gamble is talking the talk but not walking the walk.” Blumsohn also observed that P&G continued to refuse to release data to him despite being the named author on medical abstracts, draft publications and statistical reports derived from his research. P&G had previously asserted that it was “typical” for its internal statisticians to carry out analyses for supposedly independent external researchers and to withhold the full, underlying data set from the researchers. Blumsohn asserts that this makes a mockery of scientific integrity. Researchers have an ethical duty to review the data that supports their findings.
“If a company can pick and choose what data the researcher can see, that compromises the ethics of the researcher and the integrity of the researcher’s findings,” Blumsohn said at the time.
Finally, in June 2006, under continual and heavy pressure, P&G released the full research data set for its osteoporosis drug Actonel, capping a four-year quest by Blumsohn to receive it. As Blumsohn anticipated, the data fail to support P&G’s claim that Actonel is as effective as other industry-leading drugs in treating osteoporosis.
Actonel remains on the market today. While Blumsohn’s analysis of the data does not indicate it is unsafe, this episode raises serious concerns about its effectiveness, in comparison to other drugs on the market.