To Emily Dickinson, it was “fixed melancholy.” To essayist George Santayana, it was “rage spread thin.” The turns of phrase conjure different emotions, but these two writers were describing the same disorder: depression. The variance is more than a matter of literary or philosophical differences; it also reflects the fact that one was a woman, the other a man.

Therapists have long known that men and women experience mental illness differently. Yet when clinicians designed the Diagnostic and Statistical Manual of Mental Disorders, the guidebook they use to diagnose psychiatric maladies, they purposely made the disease descriptions gender-neutral. Today evidence is mounting that in turning a blind eye to gender, clinicians are doing their patients a disservice. In fact, as more researchers investigate sex differences in depression and other mental illnesses, the inescapable conclusion is that gender influences every aspect of these disorders—from the symptoms patients experience to their response to medication to the course of a disorder throughout a person's life.

Depression is the most common psychiatric disorder in the world, affecting more than 150 million people, according to the World Health Organization, or roughly 4 percent of the global adult population. In the U.S., the incidence is higher—48 million people, or 19 percent of the adult population, as reported in a recent survey conducted by the Centers for Disease Control and Prevention. (The higher U.S. figure may be linked in part to greater awareness of the disorder.)

The most obvious difference between male and female depression is prevalence. Population studies indicate that women are about twice as likely to be afflicted as men. As a result, research on depression and gender has historically focused on why women are more vulnerable to the ailment—even though, for reasons mainly to do with convenience, most studies of antidepressant drugs have recruited only male subjects.

More recently, however, researchers have started to study the deeper dissimilarities. Perhaps the most important of these, and the one most frequently misunderstood by people of both genders, is the difference in symptoms expressed by women and men. For women, the primary emotion of depression is usually sadness. For men, it is more typically anger or irritability, often coupled with recklessness. As a result, many women and men, including depressed men, mistake male depression for general frustration and restlessness rather than a serious disorder in need of intervention. Depressed men are also much less likely to seek help than depressed women, and they are much more likely to kill themselves. According to the CDC, the male-to-female suicide ratio is four to one.

The big question is whether the variations are a matter of biology or culture. Some researchers believe the brain chemistry of depression is the same in men and women but that social norms do not let men express sadness, so they often have difficulty articulating their symptoms. “They'll say, ‘I'm not getting as much done,’ or ‘I keep getting into fights with my girlfriend,’ rather than ‘I'm sad,’” says Sam V. Cochran, director of counseling at the University of Iowa and author of books on male psychology. “But once we get past that, the symptoms are pretty much the same as for the female patients.”

Cochran and others who emphasize the importance of cultural influences are increasingly in the minority. An ever growing body of evidence suggests that biology sets men and women apart in ways that have real consequences for mood and behavior—including their susceptibility to depression and other psychiatric disorders. Perhaps not surprisingly, these differences emerge from the very substances that define gender in the first place: sex hormones. Understanding the effects of these hormones on the brain may be the only way to make sure that every depressed patient gets the right treatment.

Engines of Mood
Starting in the womb and continuing through adolescence, the sex hormones, mainly testosterone and estrogen, play a leading role in brain development and, later on, in mood—and not simply in setting the stage for life's prime directive: reproduction.

Men and women make each hormone in varying amounts. Testosterone, produced in the testes, and estrogen, manufactured in the ovaries, are the most active sex hormones in men and women, respectively, but men make some estrogen and women some testosterone in their sex organs and adrenal glands. The other gender's hormone plays a vital role in men and women alike. Testosterone helps women regulate menstruation and maintain bone density, muscle mass and libido; estrogen helps men regulate fluids in their reproductive tract.

The production of sex hormones varies over a life span. Hormone levels can fluctuate from day to day and even from hour to hour. But in broad terms, output spikes first in infancy and early childhood and again in the preteen years, triggered by the hypothalamus and pituitary gland and heralding the onset of puberty. Levels of the sex hormones decline gradually from the late teens to the early or mid-50s, when women enter menopause, after which estrogen production drops sharply, and men enter a state many doctors are calling andropause, marked by a not quite so steep falloff in testosterone production. In men and women alike, researchers have linked the depleted hormone levels of old age to cognitive decline and memory loss.

The biochemistry of sex hormones in the brain is difficult to study because the hormones themselves are hard to measure and their effects are so widespread. But the evidence is strong for a primary role in gross brain physiology. The male brain tends to be larger than the female brain and matures more slowly. Although scientists have not yet pinpointed the mechanism behind the delay, animal research has shown that testosterone can increase brain size by stimulating the production of brain-derived neurotrophic factor (BDNF), a protein that contributes to neuron development. The additional growth may mean that the male brain needs more time to reach full maturity.

Much evidence also points to a relation between sex hormones and mood disorders over the course of a lifetime. Testosterone and estrogen have different effects on the brain's neurotransmitters, particularly in the hypothalamus and amygdala, both of which are involved in emotional processing. For example, studies conducted at the Albert Einstein College of Medicine in 2001 showed that during early development, testosterone and estrogen have the opposite effect on the neurotransmitter gamma-aminobutyric acid (GABA): testosterone stimulates GABA transmission, whereas estrogen inhibits it.

These polarizing effects favor first one gender and then the other. In childhood, the differences are toughest on boys. Because excess GABA is associated with seizures in infants and toddlers, estrogen's dampening effect on GABA is likely to be protective, and boys are almost twice as likely as girls to have febrile seizures. Boys also have a greater chance of developing depression during early childhood. Simon Baron-Cohen, a psychologist and director of the Autism Research Center at the University of Cambridge, argues that excesses in testosterone during the first months of brain development may make boys vulnerable to autism and other neuropsychiatric disorders. GABA, BDNF and other chemical factors that are also spurred by testosterone seem, for reasons still unknown, to be linked to these disorders. Other researchers believe that testosterone's role is indirect, making boys more sensitive to environmental stresses such as low prenatal oxygen, which in turn may produce psychiatric symptoms.

During puberty, the gender balance shifts, with girls becoming two to three times more prone to depression than boys. Researchers say surges in estrogen levels may make girls susceptible by boosting levels of cortisol, the stress hormone, and by interfering with supplies of serotonin; shortages of serotonin at that stage in life can lead to fatigue, anxiety and other symptoms of depression. For boys, testosterone may now play a protective role. In a study published in 2008 by Tracy L. Bale and her colleagues at the University of Pennsylvania, testosterone administered to female mice appeared to shield them from depressionlike symptoms, but only when given during adolescence, implying that what matters is not only which hormones the body expresses but when.

Unmasking the Symptoms
The strands of the nature-versus-nurture debate are even harder to unravel as patients enter adulthood, when symptoms of depression begin to diverge more clearly by gender. Because women are so much more likely to have the disorder and to present themselves for treatment, the diagnostic criteria for depression skew heavily toward female symptoms. The anger and restlessness typical of male depression simply do not fit the traditional definitions of the disorder, so the tests can miss men. Nor does depression fit the traditional notions of the kind of malady a man would have; for that reason, an awareness-raising campaign in 2003 sponsored by the National Institute of Mental Health (NIMH) was called “Real Men, Real Depression.”

Julie Totten of Waltham, Mass., vividly remembers the day she realized that her 54-year-old father's irritability and anger might be symptoms of depression. It was in 1990, shortly after her brother had killed himself. She had gone to the library to research possible causes of suicide and came across an article on depression in men. “When my father was well, he was social and outgoing, but he could also be very irritable and gloomy about everything,” she says. “When he was like that, you learned to tread lightly.” Her father, it seemed, was experiencing Santayana's rage spread thin.

Totten knew she would have a hard time persuading her father to see a doctor. Her brother had sought help from his family physician shortly before he died, but he complained only of external problems, such as stomachaches and weight loss. The doctor told him to eat more. Totten got her father proper care through subterfuge. She took him to his doctor when he thought he had the flu and arranged for a psychiatrist to come by during the visit. He was diagnosed with depression and put on a selective serotonin reuptake inhibitor, or SSRI, which he continues to take. Now Totten runs an organization called Families for Depression Awareness to help people recognize signs of the disorder in their loved ones and get them into treatment. She says one of the biggest support groups within the organization consists of women discussing how best to convince their husbands to seek help.

Indeed, for depressed men, seeing a doctor sooner rather than later could mean the difference between life and death. As the title of a 2008 journal article touching on the high male suicide rate put it, “Women seek help, men die.” Some researchers now advocate using rating systems designed for men, such as the Gotland Male Depression Scale, a questionnaire developed in 1999 that focuses on men's symptoms. Respondents are asked to specify, for example, the degree to which they feel irritable, restless, frustrated or aggressive.

The Medication Gap
Just as important as getting the diagnosis right is making sure men and women get treatment that fits their gender. For a long time, clinicians assumed psychiatric medications had the same effects on both sexes. But 11 years ago Susan Kornstein, a psychiatrist at Virginia Commonwealth University, published a study showing that men did not respond as well as women did to SSRIs, the class of antidepressant that includes Prozac, Zoloft and Lexapro. “It caused quite a stir,” recalls Kornstein, not to mention some red-faced defensiveness, because the clinical trials that had led to FDA approval for SSRIs had involved men only. “Researchers didn't want to deal with the difficulties of controlling for menstrual cycles,” says Kornstein, now director of the university's Institute for Women's Mental Health, so they excluded women from the trial and did not consider that the female response to the drug might be different.

But the work by Kornstein and others reveals a real gender gap for SSRI efficacy. Several recent studies suggest that these heavily prescribed medications—17 million people reported taking them between 2003 and 2006, according to the CDC—work best in the presence of estrogen. A 2008 study published in the journal Psychoneuroendocrinology found that the SSRI sertraline (Zoloft) had no effect on female rats that did not produce estrogen. The drug improved their depressionlike symptoms, however, if accompanied by estrogen treatment. When Kornstein conducted a follow-up study in 2009, she found that female patients were also more likely than male patients to experience remission after SSRI treatment, even though the depression of the female patients was on average more severe.

In contrast, Kornstein's earlier study found that men respond better to antidepressants such as imipramine (Tofranil) and buproprion (Wellbutrin) that target the neurotransmitters dopamine and norepinephrine instead of serotonin. A few years ago researchers at the NIMH and Yale University published a study that could explain why. They used positron-emission tomography to measure levels of the serotonin transporter protein—the target of SSRIs—in male and female patients who had taken antidepressants in the past but were not currently on them. Although young women showed a 22 percent reduction in the serotonin transporter in key brain regions, male patients showed no difference from healthy controls, implying that for men, depression may have less to do with serotonin deficits.

Reinforcing these results is the finding that women respond differently to antidepressants at different times of life—which may, in turn, point to why women are more likely to have depression in the first place. Kornstein found that, as with male patients, postmenopausal women did not respond as well to SSRIs as younger women did and had better results with antidepressants targeting norepinephrine and dopamine. Additionally, the Yale researchers discovered that unlike younger women (and like men), depressed postmenopausal women showed no reduction in serotonin transporter levels. These findings dovetail with the evidence in animals that SSRIs perform best in the presence of estrogen and reveal the influence of estrogen on mental health over the course of a lifetime—from the adolescent spike that permanently alters the brain's stress pathways to the loss of estrogen in menopause that has profound effects on the brain's circuitry and on women's response to medication.

The Importance of Timing
Like many women with depression, Deb Damone, 58, of Hauppauge, N.Y., first experienced what she calls “a foreboding sadness” when she hit puberty. By 17 she had been diagnosed and began taking a tricyclic antidepressant (SSRIs did not yet exist). The drug was ineffective. She avoided antidepressants until her late 30s, when a doctor put her on Prozac, which worked better. In her early 50s, though, she entered menopause, and her symptoms became worse. She was prone to mood swings and extreme sadness—Emily Dickinson's fixed melancholy—and had trouble getting out of bed in the morning. “My behavior hadn't been that erratic since I was a teenager,” she says. “I assumed I would have been in even worse shape had I not been taking Prozac.”

In fact, Damone might have had better results if she had been able to take Prozac as a young adult and then switched after menopause to a drug that targeted norepinephrine, such as venlafaxine (Effexor). Clinical psychologist Jill M. Goldstein, who studies sex differences in schizophrenia and depression at Harvard Medical School, says: “We need to put into place the notion that it takes a life-span perspective to understand these disorders.”

Goldstein has been putting her theory into practice using data from the New England Family Study, which kept medical records on thousands of children born in the region during the 1960s. The study grew out of the National Collaborative Perinatal Project, a National Institutes of Health project that followed 66,000 pregnancies in 12 cities across the U.S., starting in 1959. The researchers accumulated a trove of material, including maternal and umbilical cord blood samples and detailed medical histories of the children—all the data they might need to conduct retrospective studies on any number of conditions over the lives of the participants.

Funding for the project ran out in 1967, but the study materials are now available to researchers; it is one of the largest collections of prenatal and postnatal samples in the world. Researchers at the University of Minnesota recently used the materials to study links between childhood birthmarks and skin cancer in adults (they found a positive association). Others have drawn on the samples to investigate connections between childhood obesity and heart disease, maternal smoking and child nutrition, and lead exposure and schizophrenia.

A few years ago Goldstein, along with colleagues at Harvard and at Brown University, recruited about 1,000 of the original participants, now mostly in their 40s, for a follow-up study on depression. The researchers are looking for developmental factors linked to the disorder, many involving sex differences, with the hope of revealing hormone-driven neurochemical pathways. They are also conducting functional MRI scans to evaluate gender differences in key areas of the brain that are involved in mood, such as the hypothalamus and the amygdala. Goldstein suspects the fMRI scans of depressed women will show decreased activity in the regions of the cortex that regulate the brain's stress response.

Medicine, particularly psychiatry, has often struggled with sex stereotyping. Throughout the Victorian era and into the early 1900s, women displaying psychiatric symptoms were frequently diagnosed with hysteria, a “female” disorder no longer recognized by the medical community. As the profession shed such misogynistic views, it veered in the other direction, essentially stripping gender from diagnosis and treatment. As recently as January 2010, in a widely publicized analysis claiming that patients with mild depression do not benefit much from antidepressants, the authors could not, when asked, break out the results by gender; they did not have the data that would have allowed them to do so.

Such oversights would disappear if scientists could marshal all they are learning about the biology of depression in men and women. This knowledge could help patients find treatments more finely attuned to their body chemistry. And beyond purely medical considerations, a more sophisticated understanding of the nuances of human emotion—that depression is melancholy to some, rage to others—would deepen our knowledge of one another and of ourselves.