George Porter, a 47-year-old engineering librarian from La Canada Flintridge, first became depressed after his father's heart attack nearly seven years ago. The married father of two was overcome with sadness that wouldn't go away and lost pleasure in activities he'd once enjoyed. "I'd been a voracious reader all my life, and I found it almost impossible to get through a book," he said. He often began sobbing uncontrollably.
Porter followed his doctor's advice to see a psychologist and take medication, cycling through at least half a dozen drugs. Many helped, but none worked completely.
Although doctors have more than 20 medications to choose from when prescribing a treatment for depression, there's still little way to know which drug will work for a particular person. Many people need to try two or three drugs or drug combinations before experiencing relief. Some go through six or more. "It's a hit-or-miss, trial-and-error kind of process," said Dr. Richard A. Friedman, a professor of clinical psychiatry at Weill Cornell Medical College.
Patients have finally come to recognize depression as a treatable illness with an underlying biological cause rather than misconstruing it as a sign of weakness. Doctors are able to help more people than ever with depression simply because more people are coming to their offices for treatment. One might expect that this increase in patients would lead to a new sophistication in choosing which drug might work for a particular patient.
That's not the case.
A review article in the November 2008 issue of the Annals of Internal Medicine looked at more than 200 studies of 12 second-generation antidepressants -- primarily selective serotonin reuptake inhibitors (SSRIs) such as Prozac and Zoloft and serotonin and norepinephrine reuptake inhibitors (SNRIs) such as Effexor and Cymbalta -- and concluded that no substantial differences existed in how well they worked.
Although a more-recent review in the Lancet of the same 12 drugs concluded that certain ones worked better than others, that analysis has been criticized for reading too much into studies that are largely funded by the drugs' manufacturers.
"There's no clear evidence that one antidepressant is more effective than another," said Dr. Ian A. Cook, director of depression research at UCLA's Semel Institute for Neuroscience and Human Behavior. Even if modest differences do exist among antidepressants, he said, patients vary widely in what will work for them.
Depression is a common condition, affecting nearly 15 million Americans a year and one in six over their lifetime. The most common treatments are counseling and drugs, with a combination of the two working best.
The most effective way for a doctor to find an antidepressant that works is to look at the patient's history, because someone who has already been treated for depression will often respond to a medication that worked before. There's also a chance that someone with a family history of depression could benefit from the same drug that helped a parent or sibling.
Beyond these factors, "there is not a good way to know what medication is going to be the best for your patient," said Dr. Raymond J. DePaulo Jr., a professor of psychiatry at the Johns Hopkins University School of Medicine. Cost has become less of a concern now that most antidepressants are available in generic form for less than $20 a month, so the decision usually comes down to side effects.
Antidepressants are believed to work by blocking the reuptake of neurotransmitters such as serotonin, norepinephrine and dopamine, increasing the amount available in the synapses.
Doctors generally start by prescribing one of the SSRIs because drugs from this class are less dangerous in overdose and are least likely to cause serious side effects. Common side effects of SSRIs include nausea, weight gain and impaired sexual function. Other newer drugs include the SNRIs, which have side effects similar to those of SSRIs but may cause weight loss instead of weight gain, and the dopamine reuptake inhibitor Wellbutrin, which is less likely to cause problems with sexual function but may cause seizures.
Older drugs tend to cause more side effects. For example, monoamine oxidase inhibitors (Nardil and Parnate among them) can interact dangerously with other drugs and even some foods, and tricyclics (such as Pamelor) can increase heart rate and cause people to become dizzy when they stand. Tricyclics can also cause drowsiness, dry mouth and constipation.
Another approach is to choose a drug based on the subtype of depression.
For example, practice guidelines from the American Psychiatric Assn. suggest that people with atypical depression -- who might oversleep and overeat instead of staying up at night and losing weight -- tend to do better with SSRIs or MAO inhibitors than with tricyclics. People who have obsessive-compulsive symptoms in addition to depression may benefit from a drug used to treat both conditions, such as an SSRI. The best treatment for people with symptoms of psychosis and depression is a combination of antipsychotics and antidepressants.
Although these suggestions may help steer doctors in the right direction, Dr. Maurizio Fava, a professor of psychiatry at Harvard Medical School, cautioned against reading too much into the few studies that match subtypes to specific drugs.
"At this point, many of the treatment recommendations are oversimplifications," he said.
About 60% of patients get at least some benefit from the first drug they try, with half of those recovering fully. Doctors can add a second treatment or switch to a new one if the first drug doesn't work. The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, a large study funded by the National Institute of Mental Health that was published in 2006, confirmed that a second drug from the same class is just as likely to work as one from a different class. That is, people who don't respond to one SSRI have just as good a chance of responding to a second SSRI as to an SNRI or Wellbutrin.
Although patients become less likely to respond with each new cycle, a significant number still do. STAR*D, which looked at drugs and psychotherapy, found that 37% of patients went into remission after the first round of treatment, 31% after the second, 14% after the third, and 13% after the fourth. A third of patients in the study continued to struggle with depression after four cycles of treatment.
Compounding the problem of finding the right drug is the fact that antidepressants take so long to work. Many people, accustomed to speedy results from drugs such as aspirin, stop taking their antidepressant if they don't feel better after a week or two. STAR*D showed that it can take as long as eight weeks for a drug to begin working and up to 12 weeks to get the full effect. Doctors don't know why the drugs take so long to work; one theory is that the increase in neurotransmitters allows neurons to adapt, grow and establish new connections over time.
Friedman said that one of the most common reasons patients get incorrectly labeled "treatment-resistant" is that they haven't taken the drug for long enough or in a high-enough dose.
But waiting can be difficult for someone suffering from intense despair. As the weeks and months tick by, people with depression may be struggling with simple tasks like paying bills or getting dressed. Jobs are lost; marriages are strained. Some people kill themselves.
People who don't respond well enough to drugs and counseling still have treatment options.
One is electroshock therapy, which works well but can cause temporary memory loss. A newer alternative, called transcranial magnetic stimulation, doesn't affect memory but may be less effective. This is the treatment that Porter turned to after his long struggle with depression. He said that he was feeling much better after a month of daily treatments five days a week at UCLA.
"The critical thing for patients is not to get demoralized and give up," said Fava.