The so-called "timing hypothesis" has postulated that heightened risk might not exist for women at the younger end of the spectrum.
Originally, the WHI data was analyzed according to the number of women assigned to each arm -- whether they were taking hormones or placebos -- and did not take into account people who stopped taking the therapy or who crossed over to another arm, Liu explained.
And about 40 percent of the 16,000-plus women participating in the part of the study examining hormone use did not take their assigned treatment, a proportion that increased over time.
"We need to account for this noncompliance if we want to estimate how the effects of hormone therapy on coronary heart disease change over time," Toh said.
In their analysis, Toh and his colleagues found that, overall, women who took combined hormone therapy for the first two years of the study had more than double the risk for heart attack and other coronary problems.
The increased risk was 69 percent among those taking hormone therapy for eight years.
Among women who started hormone therapy within a decade of menopause, the risk was raised just 29 percent, not considered statistically significant. Over the first eight years of use, the risk for heart attacks was 36 percent lower in this group, the study reported.
"Our paper suggests that both timing of initiating and time since initiation are important aspects to consider when we discuss the effects of hormone therapy on heart disease risk," Toh said.
The U.S. National Women's Health Information Center has more on heart disease in women.
SOURCES: Sengwee Darren Toh, Sc.D.
All rights reserved