Spanish-speaking Latinos in the United States are more likely to begin treatment for severe mental illness in outpatient settings than are English-speaking Latinos or Caucasians. They are less likely to enter care through jail or emergency rooms. Folsom et al. (CME, p. 1173) found similar rates of hospitalization among the three groups in a large California public mental health system. Although Spanish speakers had the highest proportion of patients receiving outpatient services for major disorders, they had the fewest visits. Diagnostic and social factors may have contributed. Spanish-speaking Latinos had a higher rate of depression, and less schizophrenia and bipolar disorder. They were also less likely to have substance use disorders and more likely to live independently or with family. The English-speaking Latinos were more similar to the Caucasians. An editorial on immigration by Dr. Pedro Ruiz appears on p. 1133.

Younger patients, those with less education, and African Americans were more likely than other patients to drop out of citalopram treatment early in the first part of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. Warden et al. (p. 1189) also found that failure to achieve an early clinical response led to dropouts later in the 12-week course. A lower dropout rate was found for patients with more than one episode of depression, which suggested that they were more motivated to stay in treatment. Paddock et al. (p. 1181) have uncovered two genetic markers that were associated with response to citalopram in the same STAR*D trial. Patients who were homozygous for both marker alleles were 23% less likely to experience nonresponse to citalopram. Clinical implications are reviewed by Dr. Mary Ann Badaracco in an editorial on p. 1136. GRIK4, one of the genes involved, is described in Images in Neuroscience (p. 1148).

Parental history suggests that nonepisodic irritability and hyperarousal, or "severe mood dysregulation, " in children is not a form of bipolar disorder. Bipolar illness has a strong genetic component, and an onset during childhood is associated with a particularly strong family history. However, Brotman et al. (p. 1238) discovered that parents of children with severe mood dysregulation were much less likely to have a diagnosis of bipolar disorder, compared to parents of children with strictly defined bipolar disorder. Dr. Gabrielle Carlson discusses these findings in an editorial on p. 1140. Looking below the diagnostic level, Potash et al. (p. 1229) catalogued clinical features of bipolar disorder in two large family studies. The resulting Bipolar Disorder Phenome Database includes 5,721 subjects and allows identification of specific clinical subtypes, which may be more genetically homogeneous than the overall diagnosis. Researchers can now explore connections between clinical variables and genetics with adequate numbers of subjects to detect even moderate genetic effects.

Remission of depression in 85 patients with bipolar disorder occurred in 39% of those taking the narcolepsy drug modafinil in addition to their established mood stabilizer regimen but only 18% of those taking add-on placebo. During the 6-week study by Frye et al. (p. 1242), the modafinil group had greater reductions in depression scores at weeks 2, 4, 5, and 6. Antidepressant use did not contribute to the between-group differences. The mean modafinil dose was 174 mg/day, lower than the doses for sleep disorders, and fatigue and sleepiness levels did not change. A switch to mania occurred in similar proportions of the modafinil and placebo groups. Dr. Robert Belmaker addresses treatment of bipolar depression in an editorial on p. 1143.