Drugmaker Wyeth used ghostwriters to play up the benefits and downplay the harm of hormone replacement therapy in articles published in medical journals, U.S. researchers said on Tuesday.

Dr. Adriane Fugh-Berman of Georgetown University Medical Center in Washington and colleagues analyzed dozens of ghostwritten reviews and commentaries published in medical journals and journal supplements, many of them using documents from judicial trials.

They said Wyeth, now owned by Pfizer, paid a medical communication company called DesignWrite $25,000 to ghostwrite articles on clinical studies, including four testing low-dose Prempro, the company's combination estrogen-progestin therapy.

The team said the articles were intended to mitigate concerns that hormone replacement therapy raises the risk of breast cancer, and to support the unfounded idea that the drugs offer some protection against heart disease.

Fugh-Berman and colleagues said DesignWrite was also assigned to write 20 review articles about the drug at $20,000 each. They said the company was expected to promote unauthorized use of the drug to prevent dementia, Parkinson's disease, vision troubles and even wrinkles.

Pfizer challenged the report, noting in a statement that Fugh-Berman was a paid expert witness for plaintiffs in hormone therapy litigation.

Even with her critical perspective, she could not establish that there were inaccuracies in any of the peer-reviewed articles, or that their authors relinquished control over their work, the company said.

Use of HRT plummeted in 2002 after the publication of the Women's Health Initiative study, which found an increased risk of ovarian cancer, breast cancer, strokes and other health problems from hormone therapy.

Sales of U.S. market leader Wyeth's Prempro have fallen by about 50 percent since 2001 to around $1 billion a year.

Given the growing evidence that ghostwriting has been used to promote hormone therapy and other highly promoted drugs, the medical profession must take steps to ensure that prescribers renounce participation in ghostwriting, and to ensure that unscrupulous relationships between industry and academia are avoided rather than courted, Fugh-Berman and colleagues wrote in the Public Library of Science journal PLoS Medicine.

A 2008 study in the Journal of the American Medical Association used court papers to suggest Merck had drafted research studies for its now defunct painkiller Vioxx and then went looking for doctors to add their names to the studies before they were published.