A survey of more than 3,000 patients with symptoms of depression and anxiety found that those who combined drugs and talk therapy tended to fare better overall than those who opted for "mostly drug" or "mostly talk" therapy.

While the analysis, which appears in the October issue of Consumer Reports magazine, is based on readers' self-reports -- and therefore is not of the "gold standard" quality the medical community prefers to use as a basis for treatment decisions -- a range of experts said the carefully done survey provides a useful view of treatment outcomes with typical patients.

Patients who had combination therapy and attended 13 or more talk therapy sessions had the highest "outcome score" -- 80 -- on the 100-point scale researchers used to measure respondents' satisfaction with their treatments and improvement in their conditions. The lowest score, a 70, was for those who had "mostly talk" therapy and attended one to six sessions.

Other recently published studies had similar findings. A National Institutes of Health clinical trial of more than 400 adolescents with major depression found that combining drugs and psychotherapy was the most effective treatment. The findings, published in August in the Journal of the American Medical Association, compared cognitive-behavioral talk therapy with use of the antidepressant fluoxetine (Prozac).

Consumer Reports found that while drugs relieved symptoms of depression and anxiety faster, talk therapy was as effective as drugs when patients continued therapy for at least 13 sessions. The data also found greater incidence of side effects from medication than are cited in company-funded studies and on product labels.

"One of our big findings was that talk therapy rivaled drug therapy in effectiveness," said Donato Vaccaro, a research program leader for Consumer Reports. "People who told us that their therapy was mostly talk therapy, and it lasted for 13 sessions, had better outcomes than those whose therapy was described as mostly medication."

More than 80 percent of survey respondents reported having found a treatment that worked. The authors said people for whom treatment failed were probably less likely than others to take the 38-question survey.

The survey marks the second time Consumers Union -- an independent research and publishing group that evaluates products and services ranging from automobiles to heart-rate monitors and accepts no advertising or corporate grants -- has measured patients' mental health care experiences. It found that many more people seeking help for mental health problems are taking drugs now than were in 1995, when the previous survey results were published.

"The biggest change that we've noticed is that there has been a dramatic shift from talk therapy to drug therapy," Vaccaro said. In 1995, 40 percent of respondents reported taking medications to treat their conditions, compared with 68 percent in this survey.

Selective serotonin reuptake inhibitors (which include Prozac, Zoloft and Paxil) are now widely prescribed to treat depression and anxiety disorders, the two most common types of mental illness in the United States. But they had been on the market for less than a decade when the first survey was conducted, Vaccaro said. Since then, he said, direct-to-consumer advertising of antidepressants has increased and their use has ballooned. From 1999 to 2003, the report says, drug manufacturers spent more than $953 million advertising antidepressants directly to consumers.

About 18.8 million U.S. adults experience depression every year, according to the National Institute of Mental Health (NIMH). The three most common types are major depression (a combination of symptoms, including feelings of pessimism, hopelessness and guilt, that can disrupt daily routines and personal relationships), dysthymia (less debilitating but featuring long-term, chronic symptoms that make life less enjoyable) and bipolar disorder (mood changes from extreme highs to severe lows, also called manic-depressive illness).

Anxiety disorders -- which include panic disorder, obsessive-compulsive disorder, post-traumatic stress disorder, generalized anxiety disorder and phobias -- affect an estimated 19 million adults each year.

Clinicians said the Consumer Reports survey represents a more accurate portrayal of the patients they see in their practices than do traditional research studies. The latter typically exclude patients who have other medical conditions or don't meet certain demographic criteria. Many of those studies are funded by companies that make medications for depression and anxiety.

The survey, which was developed in consultation with academic psychiatrists and psychologists, is "much more real-world. It's a much bigger sample size," said Ari Tuckman, a psychologist with a private practice in McLean. "The problem with official research studies is that they have to limit the population they're studying. . . . It limits generalizations to the sort of people" who are studied.

But Thomas Wise, director of behavioral services for Inova Health System, said the survey is "very limited in what the individual [patient] can ascertain from it" because it is based on respondents' assessments of their conditions and doesn't include input from their therapists.

"We've got to be really careful about really taking away much from this except it's a very broad-based survey, which has merit," said Wise, a psychiatrist. "When people say they improved, it is somewhat subjective. . . . We don't really know what their problems really were."

Paulo Negro, medical director of behavioral services at Suburban Hospital, said that when interpreting the survey's results, people should remember that both types of treatment require a commitment -- to finding a medication that works and taking it regularly, or attending therapy sessions to work on problems, he said.

Negro added that while people with extreme cases of depression and anxiety disorders may need medication to get better and others find relief through attending just a few therapy sessions, "most people are in the middle. That's why it [the survey] shows the combination works better."

People filled out the survey themselves, answering questions ranging from how they were referred for mental health treatment, to which side effects they experienced from medications. They also ranked their "overall emotional state[s]" before and after they sought help, with response choices ranging from "very poor: I barely managed to deal with things," to "very good: Life was much the way I liked it to be."

More than 4,000 people responded to this year's survey. Consumer Reports based its report on the 3,079 respondents who described themselves as depressed (40 percent), anxious (17 percent) or both (43 percent). The excluded respondents had sought help for other mental health issues such as eating disorders, marriage problems and grief counseling. The magazine's 1995 survey analyzed all respondents' data.

Of the 3,079 whose responses were tallied, 2,125 visited a mental health professional; the rest sought treatment from other doctors.

The new report declares a combination approach using drugs and talk therapy the "overall winner" among the three approaches studied.

Patients who received "mostly medication" and went for six therapy visits improved as much as those on the same regimen who had 13 or more therapy visits. Talk therapy patients continued to improve with more sessions.

"If [talk therapy patients] had just six [therapy] visits or fewer, they fared worse than patients on medication," the report states. "But if they stuck with their therapy for 13 or more visits, their outcome was better than those who relied mainly on medication."

Vaccaro said the report offers valuable information for consumers faced with choosing between talk or drug therapy -- or both. Neither choice is without risk or effort. "It basically comes down to deciding between putting up with the negative side effects of a medication or just sticking out talk therapy if you can," he said.

Tuckman said a combination often works best because after receiving initial relief from symptoms via drugs, talk therapy helps patients address the root causes of their depression and anxiety.

Many respondents who took medications reported side effects, including reduced libido, weight gain and lightheadedness. For example, 53 percent of patients taking Paxil reported decreased libido and 10 percent of those on Wellbutrin reported drowsiness or disorientation. Still, 76 percent of respondents taking Paxil said the drug "helped a lot" with their symptoms, and 57 percent said Wellbutrin helped.

Some respondents either switched drugs or stopped taking them because of side effects. The survey shows the side effects were more numerous than reported in drug companies' studies. Package inserts "typically peg adverse sexual side effects as affecting 15 percent of patients or less," the report states. The Consumer Reports data show rates about three times that for Celexa, Prozac, Paxil, Zoloft and Effexor.

About 55 percent of the survey's respondents had to try two or more drugs to find one that worked for them, and 10 percent reported trying five or more medications, the report said.

Negro, a psychiatrist, said side effects are to be expected. "Every medication has some sort of side effect. You're paying a price to take the medication," he said. "But the same thing happens in talk therapy. There's a price on that. . . . You have to work because you're using tools to cause change."