NEW YORK - Supplements containing soy isoflavones may do little to preserve women's bone mass after menopause.

In a study of more than 200 women ages 46 to 65, researchers found that the soy supplement did not appear to ward off bone-density loss over 3 years. In general, women on the supplement showed the same degree of bone loss as those given a placebo -- though there was some evidence that a higher dose helped protect bone density in the hip.

The findings, reported in the American Journal of Clinical Nutrition, add to a conflicting body of research on soy and postmenopausal bone health.

Some studies have suggested that soy-based foods, isolated soy protein or isoflavone-containing supplements may be beneficial, while others have found no advantage.

Isoflavones are natural chemicals found in soybeans and certain other plant foods that are structurally similar to estrogen, and may have certain estrogen-like effects in the human body. Since declining estrogen levels after menopause spur bone-density loss, isoflavone supplements could theoretically protect women's bone mass.

The current findings, however, do not support that theory.

"I would not be able to recommend that women should take soy isoflavone tablets -- extracted from soy protein -- since these have not been demonstrated convincingly to be effective for prevention of bone loss," lead researcher Dr. D. Lee Alekel, a professor of nutrition at Iowa State University in Ames, told Reuters Health in an email.

Still, she said, women should still try to eat a variety of soy-based foods -- such as tofu, tempeh and soy-derived versions of cheese and yogurt -- since these are "nutritionally sound" choices.

For their study, Alekel and her colleagues randomly assigned 255 postmenopausal women to one of three groups: one that took 80 milligrams (mg) of a soy isoflavone supplement each day; one that took a 120-mg dose; and one given inactive placebo pills.

All of the women also took calcium and vitamin D supplements.

Over three years, women in the isoflavone and placebo groups showed similar average declines in whole-body bone mass, as well as bone density in the spine and hip area.

Women in the higher-dose isoflavone group did, however, show less bone loss in the femoral neck -- an area at the top of the thigh bone, where it meets the pelvis. But, the researchers write, because the effect was "very modest," and limited to the femoral neck, "we cannot conclude that soy isoflavones hold potential promise in the prevention of postmenopausal osteoporosis."

The idea that soy or soy isoflavones may protect bone mass comes, in part, from studies showing lower rates of hip fracture among women in Southeast Asia, where the traditional diet is rich in soy.

Those types of studies do not prove cause-and-effect, however. Because the current study looked only at soy isoflavone supplements, Alekel said it cannot speak to the potential effects of soy foods on women's bone health.