By 2050, more than two billion people over the age of 65 will inhabit the globe. In most developed countries, adults 85 years of age and older are the fastest-growing demographic group. Cognitive impairment is common among older adults, with an estimated 30%-50% prevalence of dementia among the oldest old. Many older people who do not meet criteria for dementia experience executive dysfunction, characterized by difficulty in planning and sequencing behavior or responding flexibly to the environment, leading to high levels of functional impairment. Depression is also common in late life and increases the risk of both cognitive impairment and mortality (1—3).

Despite the consequences of this demographic shift, insufficient attention has been paid to geriatric mental health issues. This may reflect the fact that few older adults seek treatment in psychiatric outpatient settings, so most mental health providers do not see many older patients in their clinical practices. This is particularly true for psychotherapists; older adults are much less likely to receive psychotherapy than are their younger counterparts, even within the Department of Veterans Affairs system, where reimbursement is not a factor (4). In addition, both psychopharmacology and psychotherapy researchers typically limit their samples to individuals under the age of 65 because of the belief that co-occurring medical or cognitive conditions common among older adults will confound the results of their investigations. The net result is that our supply of clinical experience and empirical evidence regarding psychiatric disorders in late life is limited and will be insufficient to meet what is certain to be a growing demand.

The article in this issue by Areán and colleagues (5) challenges several assumptions. Most important, their data demonstrate that psychotherapy can be effective with depressed, cognitively impaired older people. It is often assumed that people with cognitive deficits can benefit only from medications, interventions aimed at caregivers, or environmental modifications. Depressed older adults with executive dys-function, however, frequently do not respond well to antidepressant medications (6), and many do not have caregivers who can assist with interventions. The fact that 12 sessions of structured training in problem-solving skills reduced depression among individuals with executive dysfunction adds to our repertoire of effective treatments for a vulnerable, often difficult-to-treat group.

A second assumption called into question by the article, termed the "Dodo bird verdict," is that all psychotherapy approaches are equivalent (Rosenzweig [7] quoted this character from Alice's Adventures in Wonderland [8], who declares after a race that "everybody has won, and all must have prizes"). Many psychotherapists believe that specific techniques are less important than are nonspecific factors such as the therapeutic relationship. Areán and colleagues note that both problem-solving therapy and supportive therapy reduced depressive symptoms over the first 6 weeks of treatment and suggest that factors common to these therapies, such as empathy and instilling hope, may have beneficial effects for depressed older adults with executive dysfunction. At weeks 9 and 12, however, problem-solving therapy resulted in significantly greater response and remission rates than did supportive therapy, with one additional remission for every 4.5 patients treated. One proposed explanation is that the specific techniques employed in problem-solving therapy, such as setting goals, brainstorming and evaluating strategies to achieve those goals, creating action plans, and evaluating their effectiveness, have value beyond psychological support for reducing depression in individuals who have poor problem-solving skills because of their cognitive and affective disorder. According to this explanation, over the second half of treatment, the patients receiving problem-solving therapy engaged in additional practice to consolidate their problem-solving skills and as a result, continued to experience a decline in their depressive symptoms.

There is much we still need to learn about both geriatric mental health treatment and the mechanisms driving psychotherapy response. For example, given evidence suggesting a causal link between depression and cognitive impairment, is it possible that treatment of geriatric depression may improve cognition or delay the onset of dementia? Do the techniques employed in problem-solving therapy compensate for deficits associated with executive dysfunction? I have argued elsewhere that clinical trials can offer an excellent platform for conducting research on mechanisms underlying psychiatric disorders (9). Future research on treatment of depression in cognitively impaired older adults should include repeated assessment of cognitive function; in trials of problem-solving therapy, an objective assessment of everyday problem-solving skills seems particularly appropriate and would allow investigators to evaluate whether improvement in problem-solving ability in fact mediates outcomes. Incorporating neuroimaging into psychotherapy research could potentially yield valuable information about the processes underlying treatment response in individuals with executive dysfunction, who are often assumed to be poor candidates for psychotherapy because of disconnections between the frontal lobes and the limbic system. It is vital that we acquire knowledge about the mechanisms that cause and maintain cognitive and affective symptoms of geriatric depression in order to help us develop better and more effective treatments for the coming wave of older adults.