Between June 2003 and October 2004 the U.S. Food and Drug Administration (FDA) issued five warnings about increased risk of suicidality among children taking antidepressants. Whereas the first release focused on a single medication, the four that followed expanded the scope—to eight antidepressants, then to ten, and finally to all classes of these agents. Within a year of the final warning, researchers had documented 20%–30% declines in pediatric use of antidepressants. However, the focus on use led to the neglect of an important aspect of the series of FDA releases. Each warning included increasingly specific recommendations, particularly with regard to the frequency and nature of patient monitoring. In this month's lead article, Susan H. Busch, Ph.D., and colleagues report strong evidence that clinicians largely ignored FDA guidance on monitoring. Using 2001–2005 claims data for nearly 23,000 episodes of major depressive disorder in a national sample of privately insured children, they confirmed the well-documented declines in antidepressant prescribing, but they found no evidence of increased outpatient follow-up visits—even though the final FDA warning recommended weekly face-to-face contacts during the first month of treatment. The authors note that this finding "highlights the limited powers of the FDA to directly affect the care received by patients" (page 11).

As evidence of early mortality among people with severe mental illnesses has accumulated, so have efforts to detect and treat general medical conditions in this population. What degree of attention to patients' health is reasonable to expect of mental health clinicians, particularly those with little medical training? Nicholas Carson, M.D., and colleagues explored this question in a group of 47 clinicians. Transcription and coding of 120 intake sessions indicated that patients' health was discussed to varying degrees in nearly 90% of sessions, either when clinicians elicited this information (66% of sessions) or when patients volunteered it (67%). Nurses were most likely to elicit health information (100% of their intake sessions), followed by social workers (74%), psychiatrists (63%), and psychologists (44%). Clinicians most attentive to health conditions were those who used the information in case formulations (page 32). In another study focusing on patients' health, Amy M. Kilbourne, Ph.D., M.P.H., and colleagues interviewed directors of 108 of the 133 mental health programs run by the Department of Veterans Affairs to determine whether directors perceived that program clinicians or outside physicians were accountable for clinical tasks related to patients' medical conditions. Only about a third of directors (36%) reported primary accountability for monitoring diabetes and cardiovascular risk from antipsychotics, 10% reported accountability for hepatitis C screening, and 17% for obesity and weight management. Financial bonuses and co-location of general medical providers were associated with greater perceived accountability (page 38). The Taking Issue commentary raises the question of whether referral for medical evaluation is an ethical imperative. Glenn D. Grace, Ph.D., M.S., and Richard C. Christensen, M.D., M.A., argue that all mental health providers, especially those without medical training, should refer newly diagnosed patients for a thorough medical evaluation (page 3).

In the months after Hurricane Katrina, New Orleans lost 80% of its psychiatrists. Psychiatrists and other physicians relocated their families to other cities, and many had prolonged displacements or simply did not return. In the Open Forum, W. Scott Griffies, M.D., describes reasons for delays in rebuilding the city's health infrastructure and offers suggestions for policy changes that will help ensure quicker return of providers to their home communities after disasters (page 70). In a brief report, Lisa M. Brown, Ph.D., and colleagues report results of a study that evaluated mental health needs of Florida nursing home residents evacuated during hurricanes in 2004 and 2005 (page 74).

• Using data from a survey of nearly 8,500 active military personnel, Canadian researchers identified individuals who perceived a need for mental health care and why (page 50).

• This month's Law & Psychiatry column seeks to allay providers' fears that referring patients to a self-help group may result in liability claims if something harmful occurs (page 6).

• Use of signal detection analysis to reanalyze data obtained from New York City public school children after the World Trade Center attacks shows promise as a way to better target postdisaster mental health screening (page 64).