"I can remember being with friends and they were enjoying being out, but I would just as soon be home -- and they had no idea how miserable I felt," she remembered.

After her therapist agreed that an antidepressant might help, she started taking Lexapro. Immediately, her sleep improved and her spirits lifted.

"It was a miracle cure for me," she said.

But a University of Pennsylvania analysis of several studies, published in early January and widely circulated since, found antidepressants offer no more relief than placebos, also called "sugar pills," for such patients. The authors found that with few exceptions, only severely depressed patients benefit from antidepressants. They added that the impact of the medications may be "minimal or nonexistent, on average, in patients with mild or moderate symptoms."

GIVE THE MEDICINE TIME TO WORK

Psychiatrist Dr. Michael Brog of the St. Louis Analytical Institute disputes the study's findings. Brog prescribes antidepressants for 75 percent of those who frequent his Clayton office, and he's witnessed their effectiveness in treating all levels of severity.

"I'm a believer that psychotherapy and psychoanalysis are underutilized treatments," Brog (right) said. "But I also believe it's indisputable that antidepressants are effective for severe depression, chronic depression and even mild-to-moderate depression."

Brog says the new research should be taken "with many grains of salt." He sees the new study as having three critical limitations: Only six out of hundreds of studies on antidepressants were examined; the research focused on Paxil and imipramine, just two of several dozen such medications; and the course of treatment was limited to between six and 12 weeks.

It typically takes two to four weeks for antidepressants to become effective. After that, dosage adjustments or even the switch to a different medication further draw out the therapeutic process.

"To get antidepressants to work at an optimal level can take longer than 12 weeks," Brog said.

ONE SIZE DOESN'T FIT ALL

Psychiatrist Dr. J. Todd Dean, also with the institute, considers himself conservative when it comes to antidepressants. Dean rarely prescribes them in a first session, unless the client is unable to function. Those who do go on medication are encouraged to discontinue it after as few as three months.

Dean (right) finds fault with the study's failure to allow for individuality. He noted that his professional experience has not been nearly as cut and dried as that of the researchers.

"The relationship between effectiveness of medication and the diagnosis of patient is unclear," Dean said. "I've seen people with pretty serious depression respond to therapy without medication, and I've seen some with mild depression who did better with medication."

As a psychologist, Deborah Schlitt, who practices in Webster Groves, can't prescribe antidepressants. But when any client has trouble progressing in therapy sessions and completing home exercises such as journaling, she suggests the patient look into medication.

"It's hard to have the follow-through when you're depressed and you have the 'do nothings,'" Schlitt said.

Could a placebo clear up mild depression as well as medication? It's possible, but that's also held true in studies of drugs that treat physical ailments, Schlitt said.

"Placebo has been shown to work for a number of things, not just depression," Schlitt said. 

DEBATE HAS ROOTS IN WASHINGTON UNIVERSITY

The arguments over medication and other therapies in treating depression go back more than half a century with psychiatrists at Washington University playing a key role. Beginning in the 1950s, two heads of the Department of Psychiatry at Washington University, Dr. Eli Robins and Dr. Sam Guze, pioneered acceptance of what's called the "medical model" of psychiatry. In this model, physicians look for a biological cause of mental illness, rather than a psychological trigger. It marked a shift from psychoanalysis to an increased use of medication. The medical model drove a quest for effective medications to treat mental illness, including antidepressants.

"Research says most effective treatment for depression is a combination of therapy and medication," Schlitt said.

Early antidepressants had serious side effects, including drowsiness, weight gain and dizziness, as well as widespread stigma against taking them. In 1987, with the launch of Prozac, attitudes began to shift. Prozac was the first of the selective serotonin reuptake inhibitors (SSRIs). It had fewer undesirable side effects, making it easier for patients to stay on the drug. Prozac became something of a pop culture phenomenon, making the March 26, 1990, cover of Newsweek with a headline that read, "The Promise of Prozac."

That promise was followed in 1992 by the release of more SSRI medications, such as Paxil and Lexapro, which don't remain in the body as long as Prozac. Between 1996 and 2005, antidepressant use doubled in the United States. The most recent figures show that 10 percent of the population takes medication for depression, according to a study published in the August issue of Archives of General Psychiatry.

Doctors who believe in and prescribe antidepressants acknowledge the medications' drawbacks. But many don't find ineffectiveness to be one of them. The main reason most people go off antidepressants is not because they don't work: It's because patients want their sex lives back. Loss of libido, and erectile and orgasmic difficulties are typical side effects. How common are they?

"Everyone complained, 100 percent of my patients," Dean said.

"Three quarters of mine," Schlitt said.

Another possible detriment -- the flip side of antidepressants' main benefit --- is that they're designed to tone down emotional response. Often, that's a good thing. At other times, it can be disastrous. Those are the situations that doctors and their clients should be concerned with, Dean said.

"Antidepressants can make people complacent about circumstances in which they shouldn't be, like having an abusive spouse," Dean said. "What is a good outcome in that situation? Those issues are never addressed in psychological pharmaceutical literature."