Laura Ward, 41, had always attributed her excess pounds to the drugs she takes for major depression. So Ms. Ward, who is 5-foot-6 and once weighed 220 pounds, didn't try to slim down or avoid dietary pitfalls like fried chicken.
But in a clinical trial, Ms. Ward managed to lose more than 30 pounds doing low-impact aerobics three times a week. During the 18-month experiment, she was introduced to cauliflower and post-workout soreness for the first time. She and the other participants attended counseling sessions where they practiced refusing junk food and choosing smaller portions. She drank two liters of Diet Dr Pepper daily instead of eight.
Eventually, Ms. Ward, who lives in Baltimore, realized her waistline wasn't simply a drug side effect. "If it was only the medications, I would have never lost all that weight," she said.
People with serious mental illnesses, like schizophrenia, bipolar disorder or major depression, are at least 50 percent more likely to be overweight or obese than the general population. They die earlier, too, with the primary cause heart disease.
Yet diet and exercise usually take a back seat to the treatment of their illnesses. The drugs used, like antidepressants and antipsychotics, can increase appetite and weight.
"Treatment contributes to the problem of obesity," said Thomas R. Insel, the director of the National Institute of Mental Health. "Not every drug does, but that has made the problem of obesity greater in the last decade."
It has been a difficult issue for mental health experts. A 2012 review of health promotion programs for those with serious mental illness by Dartmouth researchers concluded that of 24 well-designed studies, most achieved statistically significant weight loss, but very few achieved "clinically significant weight loss."
But now a trial published online in The New England Journal of Medicine in March has provided the most comprehensive evidence yet that people with serious mental illness can lose weight, despite the challenges. Nearly 300 people with schizophrenia, bipolar disorder, schizoaffective disorder or major depression -- including Ms. Ward -- were assigned to either a control group given basic nutrition and exercise information or one whose members exercised together and attended weight-management sessions.
The mean difference between the groups at 18 months was a modest 7 pounds, but studies have shown that it is enough to reduce cardiovascular risks, the researchers noted. Nearly 38 percent of participants in the intervention group lost 5 percent or more of their initial weight, compared with only 22.7 percent of members of the control group. The difference between the groups could have been bigger, as the control group benefited from one aspect of the intervention: healthier dietary choices offered at the 10 psychiatric programs where the study took place, like baked fish instead of fried.
"This population can make a change," said Gail L. Daumit, the study's lead author and an internist at Johns Hopkins University School of Medicine. "There's been a lot of stigma that they can't do it."
Most other trials had "a narrowly defined population that excluded people with lots of co-morbidities," said Caroline Richardson at Veterans Affairs Ann Arbor Healthcare System in Michigan. But this study "applies to a lot of people."
The study suggests that weight loss may take a different trajectory for those with mental illness. Weight loss in the intervention group didn't "peak early" and then rebound a bit, as sometimes happens in programs targeted to people without mental illness, Dr. Daumit said. Instead, it "progressed over the course of the trial."
Since the study, Ms. Ward said she had regained at least 15 pounds. Still, every other day she walks for 20 minutes.
Stephen J. Bartels, a professor of psychiatry at Dartmouth and co-author of the 2012 review, said the more effective interventions for people with mental illness combined education and structured activity, focusing on both exercise and diet.
Classes and exercise programs seem to work better when they are available where mental health services are provided. And these programs should probably run six months or longer, he said.
Losing weight is challenging for anyone, let alone people with problems with executive function and memory. In Dr. Daumit's trial, researchers gave cards to carry in wallets and purses that emphasized messages like avoiding sugary drinks.
One of the few widely tried health-promotion programs for people with mental illnesses is InShape, available at 10 sites in New Hampshire and nine programs in five other states. One of its tenets is to have patients set their own goals, with the help of a health "mentor" who also sometimes accompanies them to the gym to get them past any feelings of discomfort.
In a randomized controlled yearlong intervention using InShape, to be published in Psychiatric Services this month, almost half of the 133 participants had either clinically significant weight loss (5 percent or more of body weight) or clinically significant improvements on a six-minute walk, said Dr. Bartels, the lead author.
"Many of them come to feel helpless about how they will avoid gaining weight," Ken Jue, who started InShape at Monadnock Family Services in Keene, N.H., in 2003. "We try to encourage people and say, 'You do have some control in this.' "health