Federal government says study on birth control pill's benefits was flawed
Federal officials Wednesday backed away from the findings of two major studies on birth control pills, saying the research was flawed and that a new analysis shows there is no evidence that oral contraceptives cut the risk of heart disease.
The research, presented at a medical meeting in October, created a stir because it was from the nation's largest women's health study and found that women on the pill had lower risks of heart disease and no increased risk of breast cancer. That was contrary to what many previous studies had found.
But Dr. Barbara Alving of the National Heart, Lung and Blood Institute, which funds and oversees the women's health study, said the work by researchers from Wayne State University in Detroit had not been reviewed by the study's leaders or the government before it was presented.
A new analysis by senior statisticians determined that the heart findings were flawed and that the breast cancer findings now also are suspect.
Once age and other factors were considered, they "could not find a relationship" between pill use and heart disease, said Alving.
The research was from the Women's Health Initiative, best known for its landmark finding in 2002 that taking hormones after menopause raised the risk of certain cancers and heart problems. But it is not the best kind of study for determining risks of oral contraceptives, Alving said. That's because it relies on women's memories of what drugs they used in previous years rather than actual hospital or medical records.
Previous studies that were more scientifically sound have found that pill-users have a small increased risk of blood clots, heart attacks and stroke, she said. The research presented at a medical meeting in Philadelphia two months ago reported a small benefit -- pill users reportedly had 8 percent less risk of cardiovascular disease and 7 percent less risk of developing any form of cancer.
But those findings were based on information women gave about whether they'd ever had heart problems or cancer, and were not verified with medical records as other WHI findings have been, Alving said.
"There is room for a lot of bias to be introduced," she said.
John Oliver, vice president for research at Wayne State, said the scientists were reviewing their work and would have no comment now.
"They want to look at the data. They're in contact with the Women's Health Initiative about how to proceed," he said.
The university also issued a statement apologizing for any confusion caused by the studies, which it now calls preliminary, and pledged to publish full results when they are available.
Concerns about the studies were reported by Newsday on Dec. 7 and on Saturday by The Seattle Times, which quoted disbelieving statisticians familiar with the studies.