Studies on heart disease do not provide sex-based analysis

Jan 30 2007

Heart disease differences in men and women continue to be poorly understood because women are included in clinical trials far less than men, and even when women are included, study results are not reported by sex, according to a study in the current issue of Mayo Clinic Proceedings.

The study shows that three-fourths of clinical cardiovascular trials published in leading general medical and cardiology journals during the last six months of 2004 did not provide sex-based analysis.

In a review of 645 cardiovascular clinical trials published from July 1 through Dec. 31, 2004, only 153 provided sex-specific reporting -- defined as reporting results for women and men in a format that allows data to be specifically extracted for each sex. In addition, the authors found that 7 percent of the studies did not report the participants' sex, and 3 percent included no women, despite studying conditions that affect both sexes.

"Heart disease is the No. 1 threat to a woman's health and we need to be able to tell women whether the diagnostic tests we order are accurate and how treatments will affect them, but today we don't have enough data specific to women," says Sharonne Hayes, M.D., an author of the collaborative study and director of Mayo Clinic's Women Heart Clinic. "We hope this analysis will drive the behavior of researchers. If more women are included in trials and the results are reported by sex, it will help physicians provide the best care possible to both men and women."

Policy changes in 1986 by the National Institutes of Health (NIH) were aimed at increasing participation by women and minorities in research. Little progress was made until 1993, however, when NIH-funded research required by law that women and minorities be included in trials unless a clear reason was given for exclusion. The law required inclusion of women and minorities but did not require reporting the data. The authors of the Proceedings report found that clinical trials funded by the NIH were more likely to report results by sex than non-NIH-funded trials (51 percent versus 22 percent). While sex-based reporting has increased as a result of the NIH requirement, the authors challenge journal editors to create a similar policy of reporting sex-based results, regardless of the funding source.

"A change by the NIH and other sources of funding to encourage sex-specific research is just the beginning," says Mary Norine Walsh, M.D., another study author and a cardiologist at The Care Group in Indianapolis. "More sustained change will occur when those planning large clinical trials include enrollment of enough women to allow for pre-specified endpoint analysis, and when journal editors and reviewers uniformly require such analysis."

Physicians have gained important information that has improved patient care through cardiovascular trials that reported sex-based results, the authors say. For example, because men and women have differences in their hearts' electrical systems, women are at greater risk of sudden death when given several commonly used antibiotics and anti-arrhythmic drugs, and some diagnostic tests need sex-specific interpretation. These data help shape gender-specific care plans.

In addition to challenging researchers to include men and women in trials and report results by sex, the authors say patients also can play a role.

"When a physician recommends a certain treatment or test to a woman, she should ask, 'Were women included in the research?'" Dr. Hayes says. "When I tell a patient, 'You need this medication or test or procedure,' I should be able to say, 'This has been tested in women.' In many cases now, I cannot say that. But I think those questions from women can drive our behavior and push us to make sure future research is more applicable to them."

Dr. Hayes also encourages women to consider enrolling in a clinical trial. Participation in many trials is very low risk and requires no more effort than giving blood or allowing researchers to use data from a particular test.