Every day, millions of Americans take antidepressant medications prescribed by their doctors to alleviate their symptoms and help them feel better.

More than 30 years of research has shown the drugs to be safe and effective. But a pair of recent Penn studies have demonstrated the drugs may not work the same for everyone. The groundbreaking work originating from the lab of Robert J. DeRubeis, a Penn professor of psychology, is generating much debate and could lead to major changes in the way doctors treat mild and moderately depressed patients.

DeRubeis has been studying depression since the 1980s, when the first selective serotonin reuptake inhibitors, or SSRIs, came to the market. Back then, he says, the emergence of SSRIs caused heated debates about the effectiveness of psychotherapy.

“There was a recognition that [depression] is a disorder that could be understood better and better, and where treatments could be made more effective,” he says. “There was also an assumption that the medications are simply effective and we don’t need to know more.”

The ability of antidepressant medications to reduce depressive symptoms varies considerably, depending on the severity of the symptoms when treatment is initiated. Numerous previous studies have demonstrated the effectiveness of these drugs in treating those with very severe depression. DeRubeis’ most recent study affirms those findings.

“For very severe depressions, the benefits of medications are clear and substantial,” says DeRubeis.

Yet these patients are estimated to make up only half of the millions of people around the world who are prescribed and take the mood-enhancing drugs. For the majority of people whose depression is considered to be mild, moderate, or even somewhat severe, little research exists to prove the benefit of antidepressant medication compared with those who are given a placebo, or sugar pill.

For the first time, the DeRubeis team sought to test whether or not SSRIs actually help alleviate milder forms of depression.Looking at 30 years of antidepressant-drug treatment data, they discovered the majority of patients who were prescribed antidepressants did not suffer from a depression severe enough to be improved by the medications.

“There is little evidence that the medications yield specific benefit, beyond what is provided by engagement in treatment and the resulting boost in the patient’s expectation for improvement,” says DeRubeis, the Samuel H. Preston Term Professor in the Social Sciences and a professor of psychology in the School of Arts and Sciences.

The implicit conclusion is that doctors should pause before writing a script for Prozac to patients newly diagnosed with a lesser form of depression. Many studies have shown these patients respond well when given a placebo combined with psychotherapy.

DeRubeis says for too long, physicians may have prescribed antidepressants with a misunderstanding that research was done at all levels of depression. Confronted with the reality that there is a lack of evidence for specific benefits of the medications, he says doctors should rethink their ways and compare the expected costs and benefits of the medications against proven alternatives such as self-help, exercise, or psychological therapy.

The findings could also serve to motivate mental health patients to take a more active role in their care, he adds.

“I’m not saying we need to change entirely how the practice of prescribing these medications should occur. But if I was reading this stuff for the first time, I’d say, ‘Are you kidding me?’ I would want to know that there would be more research,” he says. “That’s the issue; we need more data.”

DeRubeis’ team analyzed data from studies that spanned a 30-year period to gauge the benefit of antidepressant medication compared to placebo pills across a range of patients with varying degrees of depression. The team collected data from placebo-controlled trials of antidepressants approved by the U.S. Food and Drug Administration for the treatment of Major or Minor Depressive Disorder. Studies comprised of adult outpatients and included a medication versus placebo comparison for at least six weeks.

Researchers calculated depression severity scores before and after treatment using the standard Hamilton Rating Scale for Depression. The scientists learned that medication versus placebo differences varied substantially, relative to the severity of depression suffered by patients. Those patients shown to have less severe depression experienced little to no greater improvement in their symptoms from antidepressant medication as compared to a placebo sugar pill.

The paper, published earlier this month in the Journal of the American Medical Association, has touched off a fierce scholarly debate over the methods DeRubeis’ lab used to construct the analysis.

In blogs and prominent newspaper columns alike, critics are questioning why the team narrowed its focus to include only two antidepressants when there are 25-plus others available to consumers. A decision to exclude data from larger clinical studies in which scientists controlled for the placebo effect also has drawn fire.

DeRubeis contends such clinical trials are tantamount to “studying a group of people who can’t be replicated out in the world.” They were excluded, he says, because the data they generate is skewed. And there is no evidence that any one drug is superior to another in the treatment of minor to severe depression.

Despite the national attention the paper has drawn, DeRubeis insists a study he published last month that showed how antidepressant medications may change patients’ personalities “was actually one of the most fascinating findings I’ve been a part of.”

Scientists previously assumed personality changes observed in patients taking SSRIs for depression were the byproduct of alleviating depressive symptoms. But in a study of 240 participants with Major Depressive Disorder, researchers found changes in neuroticism and in extraversion were far greater for patients taking the drug paroxetine, relative to those receiving placebo, with cognitive therapy in between.

“Something happened to these people that is very difficult to figure out. They went on to be resistant to relapse.” Not too unlike Bill Murray’s character in the movie “Groundhog Day,” participants “have a new way to be in the world they never had experienced before,” DeRubeis says.

The study, published in the December issue of Archives of General Psychiatry, is the first of its kind to look carefully at other potentially important effects of SSRIs on the psychology of the patient. DeRubeis’ hope is it will lead to more effective therapies.

“We’ve not been, as a field, interested in finding out what these medications do to people. We know they work and we know the side effects,” he says. “It’s almost as if we assume these medications go in and attack depression when in fact, if our findings are trustworthy, it would suggest some permanent or semi-permanent change in the brain is taking place. That’s where the research is heading.”