While only 41 percent of studies reviewed by the FDA for approval made some mention of gender bias, 26 percent of those that did identified some difference in safety or effectiveness by sex, supporting the importance of such analyses, Redberg's group noted.
The incomplete safety and efficacy data for women "may mean that more women are experiencing adverse events for a device lacking a mortality benefit," the researchers warned.
They pointed to implantable cardioverter-defibrillator registry data showing higher likelihood of in-hospital and major adverse events compared with men but no mortality benefit in primary prevention for women with heart failure to compensate.
The justifications given for the failure to adequately address gender just perpetuate the status quo, Redberg's group added.
For example, they noted one heart failure study that enrolled only 22 percent of women despite equal representation of women among Medicare patients with impaired left ventricular function, which gave as the explanation that these demographics were typical for a U.S. device trial.
"There is no question that women continue to be underrepresented in clinical trials," said Dr. Michelle O'Donoghue, an investigator in the TIMI study group in the cardiovascular division of Brigham and Women's Hospital. "This is true despite the fact that cardiovascular disease remains the leading cause of death in women in the United States."
Despite the discrepancy, O'Donoghue said it's unclear why female participation lags behind men. It is not known if women are approached less frequently than men or women are less willing to participate.
The National Institutes of Health and the FDA have been calling for inclusion of women in trials since the mid-'80s, including a specific directive from the FDA in 1994 requiring that device approval applications address the ratio of men and women in the studies and differences in safety and effectiveness by gender.
The researchers systematically reviewed all 78 such data summaries released to the public for class III cardiovascular devices approved from 2000 through 2007.
Men Dominate Heart Device Trials
Men accounted for 67 percent of patients in the studies included in the FDA's Summaries of Safety and Effectiveness Data, ranging from 28 to 92 percent across the studies but without any increase in the proportion of women across the eight-year span.
Nor did the proportion with analyses by sex or comment on gender bias improve over the review period.
But despite the strong statements, Dr. Christopher Cannon, associate professor of medicine at Harvard University, said he was not sure what to make of the study.
"When patients are enrolled, they are randomly assigned to treatment, so that there would not be any bias," Cannon said. "If they are saying it is biased because only one third of patients were women and not half, that seems to not be the right word to describe it.
"Underrepresented would perhaps be better," Cannon continued.
Of the 123 studies, 28 percent did not even have the proportions of men and women reported. Of those that did, almost half didn't report the proportion for all the patients enrolled, "which suggests an important source of missing data," according to Redberg and the other researchers.
"Raw data or statistical analyses rarely were provided, and sex-specific analyses of study primary end points were even less common," they added in the paper.
Resolving this situation may take further action by the FDA, Redberg and colleagues suggested.
The agency has been working on a guidance document for study and analysis of gender differences in cardiovascular device trials; an act pending in Congress could go even further in mandating an end to the disparities, they noted.
"As we move forward, I agree that it is critical to encourage the enrollment of women in clinical trials," O'Donoghue said. "The key is that we don't want women to be deprived of therapies with proven benefit, but rather to find new ways of identifying those women and men who benefit the most."