A spate of studies since then—not only of widows but of parents who had lost a child, tsunami survivors and others—has further confirmed and refined that initial description. In 2008 researchers got their first hint of what complicated grief disorder looks like at the neurological level. Mary-Frances O’Connor of U.C.L.A. scanned the brains of women who had lost their mother or a sister to cancer within the past five years. She compared the results of women who had displayed typical grief with those suffering from prolonged, unabated mourning. When, while inside the scanner, the study participants looked at images of the deceased or words associated with the death, both groups showed a burst of activity in neurological circuits known to be involved in pain. The women with prolonged grief, however, also showed a unique neural signature: increased activity in a nub of tissue called the nucleus accumbens. This area, part of the brain’s reward center, also lights up on imaging scans when addicts look at photographs of drug paraphernalia and when mothers see pictures of their newborn infant. That does not mean that the women were addicted to their feelings of grief but rather that they still felt actively attached to the deceased. Meanwhile clinical studies have shown that a combination of cognitive therapy approaches used to treat major depression and post-traumatic stress may help some people with complicated grief work through it.
As these and other studies began to pile up, a few researchers turned to complex statistical analysis to validate more precisely the exact combination of features that define the condition. In 2009, more than 10 years after the Pittsburgh panel, Prigerson published data collected from nearly 300 grievers she had followed for more than two years. By analyzing which of some two dozen psychological symptoms tend to cluster together in these participants, she devised the criteria for complicated grief: the mandatory presence of daily yearning plus five out of nine other symptoms for longer than six months after a death [see box at right]. This is exactly the type of rigorous, quantitative study that is needed before a condition makes it into the DSM. “People who meet the criteria for complicated grief do not necessarily meet criteria for either depression or post-traumatic stress disorder,” says Katherine Shear, a professor of psychiatry at Columbia University. “If you didn’t have this disorder [in the DSM], then those people would not get treatment at all.”
The case for diagnosing people as depressed and treating them accordingly when they are still newly bereaved is more contentious. Although some symptoms of grief and depression overlap (sadness, insomnia), the two conditions are thought to be distinct. Grief is tied to a particular event, for example, whereas the origins of a bout of clinical depression are often more obscure. Antidepressants do not ease the longing for the deceased that grievers feel. So in most cases, treating grieving people for depression is ineffective.
A few studies, however, have suggested that mourning may trigger depression in the same way that other major stresses—such as being raped or losing one’s job—can bring about the condition. If so, some people who grieve may also be clinically depressed. It seems unfair, advocates of changing the DSM argue, to make mourners wait so long for medical help when anyone else can be treated for depression after just two weeks of consistent depression. “On the basis of scientific evidence, they’re just like anybody else with depression,” says Kenneth S. Kendler, a member of the DSM-5 Mood Disorder Work Group, which reviews all proposed changes to the manual related to anxiety, depression and bipolar disorder (a condition characterized by extreme mood swings). It is for this reason that the group recently suggested deleting the clause that specifies a two-month wait before mourners can receive a diagnosis of, and therefore treatment for, depression.