Diabetics at high risk for heart attacks can lower their odds of heart trouble by eating the ultra-low-fat diet created by cardiologist Dean Ornish, doing moderate exercise and reducing stress, suggests a study that will be released Friday.

The first large research on how the Ornish program affects diabetics will be reported by psychologist Michael Sumner of the Preventive Medicine Research Institute, Ornish's center in Sausalito, Calif., at the American Psychosomatic Society meeting in Vancouver, B.C.

The study included 461 diabetics who had heart disease or at least two risk factors, such as high blood pressure and cholesterol. Here's the program they followed:

A vegetarian diet with 10% of calories from fat that emphasizes whole grains, fruits and vegetables. The average American eats a diet that's 30% to 40% fat.
At least three hours of aerobic exercise a week. Most walked briskly.
An hour a day of stress management. Participants learned yoga, relaxation exercises and meditation.
Two-hour sessions twice a week in groups offering health education and mutual support.
After 12 weeks, participants lost an average of 12 pounds, Sumner says. Their blood sugar levels, blood pressure and LDL (bad cholesterol) fell significantly. Also, 39% could reduce their diabetic medication or stop taking it, he says. There's evidence that several of these changes, such as lower LDL levels, reduce the risk of future heart attacks, Sumner adds.

Study participants will be tracked for up to three years to check progress, he says.

But most diabetics aren't willing to change their lives as drastically as study participants did, says Roberta Anding, a registered dietician in Houston and certified diabetes educator with 25 years of experience.

"You've got a 50-year-old man who's always eaten white bread, and now I need to get him to eat wheat bread," she says. "Some don't want to make any changes; some will meet me halfway.

"It takes a highly motivated person to get to 10% fat, and for them this would work."

The absence of a randomly chosen control group that received no unusual care makes the study inconclusive, says diabetes physician Robin Goland, co-director of the Naomi Berry Diabetes Center at Columbia University Medical Center in New York.

"Anything that helps people to lose weight, to eat less and exercise more is good for diabetes care," she says. But plans need to be tailored to patients, because people metabolize food differently.

"There's never going to be a one-size-fits-all diet. And my patients need plans they can stick to for life. This is a marathon, not a sprint."